Journal of Environmental Management 143 (2014) 99e105
Contents lists available at ScienceDirect
Journal of Environmental Management journal homepage: www.elsevier.com/locate/jenvman
Cysteine-grafted nonwoven geotextile: A new and efficient material for heavy metals sorption e Part B M. Vandenbossche a, H. Vezin b, N. Touati b, M. Jimenez a, M. Casetta a, *, M. Traisnel a a Unité Matériaux et Transformations (UMET), Ingénierie des Systèmes Polymères (ISP), CNRS-UMR 8207, ENSCL, Université Lille Nord de France, 59652 Villeneuve d’Ascq Cedex, France b Laboratoire de Spectrochimie Infrarouge et Raman (LASIR), UMR-CNRS 8516, Université Lille Nord de France, Bâtiment C5, 59655 Villeneuve d’Ascq Cedex, France
a r t i c l e i n f o
a b s t r a c t
Article history: Received 14 January 2014 Received in revised form 1 April 2014 Accepted 2 May 2014 Available online
The development of a new material designed to trap heavy metals from sediments or wastewater, based on a polypropylene non-woven covalently grafted with cysteine, has been reported in a previous paper (Part A). The non-woven was first functionalized with acrylic acid (AA) which is used as spacer, and then cysteine was immobilized on the substrate through covalent coupling in order to obtain the so-called PP-g-AA-cysteine. Some preliminary heavy metals adsorption tests gave interesting results: at 20 C for 24 h and in a 1000 mg/L heavy metals solution, PP-g-AA-cysteine adsorbs 95 mg Cu/g PP (CuSO4 solution), 104 mg Cu/g PP (Cu(NO3)2 solution), 135 mg Pb/g PP (Pb(NO3)2 solution) and 21 mg Cr/g PP (Cr(NO3)3 solution). In this second part of the work, heavy metals sorption tests were carried out with Cu (II), Pb (II), and Cr (III) separately, in order to determine the sorption capacity of this new sorbent as a function of (i) the heavy metals concentration in the solution, (ii) the contact time with the solution, (iii) the pH and (iv) the ionic strength of the solution containing heavy metals. Moreover, the sorption capacity of PP-g-AA-Cysteine was studied using a polluted solution consisting of a mixture of these different heavy metals. An Electron Paramagnetic Resonance study was finally carried out in order to determine the coordination geometry in the environment of the copper trapped by the PP-g-AAcysteine. Ó 2014 Elsevier Ltd. All rights reserved.
Keywords: Adsorption Heavy metals EPR Polypropylene Remediation
1. Introduction Currently, the accumulation of massive amounts of sediments leads to important economic and environmental issues. In order to maintain the depth of the navigational waterways, harbors and estuaries worldwide (Stronkhorst et al., 2003), and also to limit the remobilization of contaminants which are present in sediments (Eggleton and Thomas, 2004), dredging operations occur regularly. But most of these dredged sediments can be considered as waste as they contain harmful components such as organic compounds and heavy metals (Singh et al., 1998). Because of the high concentration in heavy metals such as Arsenic, Cadmium, Chromium, Copper, Mercury, Nickel, Lead, and Zinc (Gouzy and Ducos, 2008), these sediments must be put in storage centers. But as storage centers are
* Corresponding author. E-mail address: [email protected] (M. Casetta). http://dx.doi.org/10.1016/j.jenvman.2014.05.002 0301-4797/Ó 2014 Elsevier Ltd. All rights reserved.
saturated, sediments have to be treated in order to remove heavy metals and to give them a second life. Geotextiles are commonly used in filtration and drainage. The grafting of specific biomolecules provides heavy metals adsorption properties to the textile. Cysteine was chosen as a great number of chelating biomolecules (such as glutathione, phytochelatins and metallothioneins) contain a large amount of cysteine units (Cobbett, 2000; Klaassen et al., 1999; Maret et al., 1997; Zenk, 1996). The grafting of cysteine on polypropylene nonwovens (PP) was already carried out and optimized in our group and is detailed in the Part A of this study (Vandenbossche et al., 2014). In the present paper, a heavy metals sorption investigation is proposed, in order to determine the capacity of this new material, PP-g-AA-cysteine for short, to trap copper (II), lead (II), and chromium (III) in various experimental conditions, i.e. when initial metal concentration, contact time, pH and ionic strength vary. Electron Paramagnetic Resonance (EPR) analyses were also carried out on copper solutions and on PP-g-AA-cysteine e copper in order to determine the heteroatoms involved in the copper sorption.
100
M. Vandenbossche et al. / Journal of Environmental Management 143 (2014) 99e105
2. Materials and methods
Thermo Scientific and Mono-elements coded hollow-cathodes lamp for Pb, Thermo Scientific).
2.1. Preparation of the optimized PP-g-AA-cysteine Squares (5*5 cm2) were cut from polypropylene (PP) nonwoven INTN50 (50 g/m2, provided by PGI nonwovens, France), washed and dried as already described in a previous paper (Vandenbossche et al., 2013). Acrylic acid (purity 99.5%, provided by Acros Organics) was graft-polymerized on PP by a cold plasma process (PPg-AA) also previously optimized and reported (Vandenbossche et al. 2013). Then cysteine (purity 97%, provided by Sigmae Aldrich) was immobilized on PP-g-AA by a chemical coupling between the carboxylic acid group of acrylic acid and the amine group of cysteine using a carbodiimide: N-(3-dimethylaminopropyl)-N0 ethylcarbodiimide hydrochloride (purity 98%, provided by SigmaeAldrich). The immobilization of cysteine was carried out according to the optimized conditions determined in the Part A (Vandenbossche et al., 2014): PP-g-AA was first immersed in a solution containing the carbodiimide (1 h at 4 C), and then immersed in a 0.229 mol/L cysteine solution for 28 h at 20 C. Samples were then washed for 1 h in an ultrasonic bath, and for 6 h in a Soxhlet extractor with distilled water. Finally, PP-g-AA-cysteine samples were dried under vacuum for 16 h.
2.3.2. Preparation of the solution for the calibration of the atomic absorption spectrometer The determination of the concentration of heavy metals in the solution obtained after the digestion of the metals-chelated-textile was carried out by the standard addition method: the analysis was first made with a blank, and then with the solution to analyze of known dilution, and then with the same solution in which 1, 2, or 3 mg/L of heavy metals were added. 2.3.3. Influence of the metal concentration Cysteine-grafted-polypropylene samples were immersed in 100 mL of 100 mg/L, 250 mg/L, 500 mg/L, 1000 mg/L, and 5000 mg/L contaminated solutions at 20 C and pH ¼ 4.5 for 24 h. 2.3.4. Kinetic studies Cysteine-grafted polypropylene samples were immersed in 100 mL of 1000 mg/L contaminated solutions (Cu II, Pb II, Cr III) at 20 C and pH 4.5. For each heavy metal, the amount of metal trapped by the functionalized textile was determined as a function of time. It was then possible to determine the kinetic law of adsorption followed by the grafted surface.
2.2. Preparation of the artificially contaminated solutions Some artificially polluted solutions were prepared in order to determine the adsorption behavior of the PP-g-AA-cysteine as a function of the oxidation state, the ionic radius of the metal and the counter-ion. 1000 mg/L solutions of copper sulfate, copper nitrate, lead nitrate and chromium nitrate were prepared in a 1 L-volumetric flask, and the 5000 mg/L solutions were prepared in a 100 mL-volumetric flask. From the solutions containing 1000 mg/L heavy metals, various contaminated solutions were prepared, with the following concentrations: 100 mg/L, 250 mg/L, and 500 mg/L. In order to prepare the solutions containing 1000 mg/L and 5000 mg/L of heavy metals, copper sulfate solution was diluted in distilled water (conductivity ¼ 0.2 mS/cm) while copper nitrate, lead nitrate and chromium nitrate were dissolved (see supporting data, table S1) using these initial reagents: 0.1 M CuSO4.5H2O (provided by Sigma Aldrich), Cu(NO3)2.xH2O (purity ¼ 99.999%, provided by SigmaeAldrich), Pb(NO3)2 (purity 99.0%, provided by SigmaeAldrich), Cr(NO3)3 (purity ¼ 99.0%, provided by Sigmae Aldrich). Finally, a solution containing 1000 mg/L copper, 1000 mg/L chromium and 1000 mg/L lead was prepared by dissolving respectively 2.951 g of Cu(NO3)2, 7.700 g of Cr(NO3)3 and 1.600 g of Pb(NO3)2 in distilled water in a 1 L-volumetric flask. 2.3. Sorption studies 2.3.1. Chelation of heavy metals and sorption studies One cysteine-grafted polypropylene sample was immersed for 24 h in 100 mL of the previously described heavy metal solution thermostated at 20 C. The sample was then washed in a bath of 50 mL of ultrapure water (conductivity ¼ 0.055 mS/cm) in order to remove heavy metals located at the surface which did not interact with cysteine or acrylic acid. Then the textile was digested with sulfuric acid at 95% and hydrochloric acid at 37% (provided by VWR BDH Prolabo) to remove heavy metals which interacted with the surface. Ultrapure water was added to the solution, and the concentration of the diluted heavy metal containing solution was determined using flame atomic absorption spectrometry (Thermo Solaar S4 AA Spectrometer, Thermo S Series, Multi-elements combined coded hollow-cathode lamp for CreCueMneNi,
2.3.5. Influence of pH Cysteine-grafted-polypropylene samples were immersed for 24 h in 100 mL of 1000 mg/L contaminated solutions at 20 C. 1M HCl or 1M NaOH was added in order to reach the following pH: 2.7, 4.5, 6.0, or 8.0. 2.3.6. Influence of the sodium concentration Cysteine-grafted-polypropylene samples were immersed in 100 mL of 1000 mg/L contaminated solutions in which NaCl (Sodium chloride for analysis EMSUREÒ ACS, ISO, Reag. Ph Eur, Merck) was added to reach the following concentrations: 10, 20, or 30 g/L. These experiments were carried out at 20 C and pH ¼ 4.5. 2.3.7. Influence of a mixture of heavy metals in the solution Cysteine-grafted-polypropylene samples were immersed in 100 mL of an artificially contaminated solution containing 1000 mg/L of copper, 1000 mg/L of lead and 1000 mg/L of chromium (for all these heavy metals the counter-ion was NO 3 ) at pH ¼ 4.5, 6, 7 and 8 in presence of 30 g/L NaCl or without NaCl. 2.4. Physical characterizations of the grafted textile 2.4.1. Langmuir isotherm According to the classification of Brunauer et al. (1940), the Langmuir isotherm is the first physical model of sorption (type I). It is an empirical sorption model (Foo and Hameed, 2010) which characterizes the adsorption of a monolayer of compounds on a microporous material. In this model, (i) the energy is assumed to be the same all over the surface, (ii) the trapped molecules are assumed not to interact together or migrate over the surface and (iii) at the maximum of sorption, only one monolayer can be formed on the free surface. In the case of the adsorption on a solid surface of a molecule from a solution, the Langmuir isotherm model can be expressed according to the following equation (Langmuir, 1918):
q ¼ qm *b*C=ð1 þ b*CÞ
(1)
where q (mg/g PP) is the sorption capacity, qm (mg/g PP) is the maximum sorption capacity, C (mol/L) is the concentration of heavy metal in the solution, and b (L/mol) is the Langmuir constant. In this
M. Vandenbossche et al. / Journal of Environmental Management 143 (2014) 99e105
101
case, the free Gibbs energy can be calculated with the following formula:
homogeneous sphere (Cooney, 1998). The HSDM equation has the following form:
DG0 ¼ RTlnðbÞ
. . . vq=vt ¼ DS r2 v r2 vq vr vr
(2)
where R (8.31 J/mol/K) is the perfect gas constant, and T (K) is the temperature.
2.4.2. Kinetic studies: pseudo-first-order and pseudo-second-order The adsorption mechanism is assumed to occur in three steps (Chiou and Li, 2003): first, the molecule moves through the solution, then from the solution towards the surface which corresponds to the diffusional step that can be divided into two stages (1) diffusion across the liquid film surrounding the textile (external diffusion or film diffusion) and (2) diffusion in the liquid contained in the pores and/or along the pore walls (internal diffusion or intraparticle diffusion) (Qiu et al., 2009); and in a third step, the molecule is trapped at the surface of the solid material which corresponds to the adsorption onto active sites. Literature listed a lot of kinetic laws allowing the evaluation of the adsorbents performance and giving an idea concerning the implicit mechanism (Chatterjee and Schiewer, 2014; Chu, 2002). In this paper, some kinetic laws were chosen because of their efficiency to describe adsorption of compounds from aqueous media onto solid surfaces: the intraparticle diffusion models, homogeneous solid diffusion model (HSDM) and linear driving force (LDF) (Qiu et al., 2009) which are mass transfer modeling (MTM) and also the two well-known kinetic models used in this field the pseudo-first-order (PFO) and pseudo-second-order (PSO) equations, which are surface reaction modeling (SRM) (Gerente et al., 2007). As it was reported by Chatterjee and Schiewer, the intrinsic sorption reaction is assumed to be fast in MTM, and thus, the diffusion processes are considered as rate limiting steps, whereas in SRM, the mass transfer is assumed to be fast and the trapping of the adsorbent is the rate limiting step (Chatterjee and Schiewer, 2014). The first equation of the sorption capacity in a solid/liquid system, for example the sorption of heavy metals in solution on a material surface, is called the Lagergren rate equation or pseudo-first-order. This kinetic model has the following form (Gerente et al., 2007):
dqt =dt ¼ k1 ðqe qt Þ
(3)
where qe and qt (mg/g) are respectively the sorption capacities at equilibrium and at time t, and k1 is the pseudo-first-order sorption rate constant. The pseudo-first-order equation in linear form is the following one:
lnðqe qt Þ ¼ lnðqe Þ k1 t
(4)
However, the pseudo-first-order equation cannot explain all the kinetic phenomena, and it is the reason why other models have been developed. Generally, the pseudo-second-order model is much more successful, and has the following form (Gerente et al., 2007):
dqt =dt ¼ k2 ðqe qt Þ2
(5)
(7)
where DS is the intraparticle diffusion coefficient, r the average particle radius, q the adsorption capacity and t the time. An exact solution can be given for the “infinite bath” case where the system is considered free of adsorbate and the adsorbate concentration at the surface remains constant (Qiu et al., 2009). Some approximations can be carried out in order to simplify the equation. Indeed, particles are considered spherical at any particular time and the external film resistance can be neglected according to the constant surface concentration (Cooney, 1998). Thus, two solutions are considered depending on the sorption time. For a short time (when qa/qN < 0.3 with qa the average value of q, and qN the average concentration in the solid at infinite time), the equation can be simplified as:
1=2 . qa =qN ¼ 6* DS R2 *p *t 1=2
(8)
where R is the total particle radius. For a long time, the equation is simplified as:
. . qa =qN ¼ 1 6 p2 exp DS *p2 *t R2
(9)
Equation (8) assumes that adsorption rate decreases along with the increasing adsorbent particle size and vice versa (Qiu et al., 2009). Concerning Equation (9), the assumption of constant surface concentration for HSDM is not likely to be verified at a long time, and consequently, only short time experiments are considered (Cooney, 1998). Intraparticle diffusion can also be modeled using the linear driving force (LDF) approximation. This model assumes that the rate of mass transfer linearly depends on the concentration difference between the external surface of the adsorbent and the solution. The following equation corresponds to this model:
dq=dt ¼ kS *S0 *ðqi qÞ
(10)
where S0 is the particle surface area per particle volume and kS the transfer coefficient. 2.4.3. Electronic paramagnetic resonance spectroscopy EPR spectroscopy experiments were performed at X band using a Brüker ELEXSYS E500 spectrometer in continuous waves (CW). The spectra were recorded on TM cavity with 10 mW of microwave power and with modulation amplitude of 4 G. The spectra were collected at room temperature on the grafted polypropylene directly inserted in a 8 mm quartz tube. The g-factor, which is a constant of proportionality, was determined in order to have information on the properties of the electron in a certain environment. When there is a hyperfine structure, it is possible to determine the hyperfine constant A which corresponds to the distance between two consecutive stripes.
where qe and qt (mg/g) are the sorption capacities at equilibrium and at time t respectively, and k2 is the pseudo-second-order sorption rate constant. The pseudo-second-order equation in linear form is as follows:
3. Results and discussion
1=ðqe qt Þ ¼ k2 t þ 1=qe
3.1.1. Sorption capacity of the material The sorption capacity of the textile depends on the heavy metals concentration of the artificially contaminated solution. Fig. 1 (and Figure S1) shows the evolution of the amount of heavy metals
(6)
The homogeneous solid diffusion model (HSDM) is a typical intraparticle diffusion model which can describe mass transfer in a
3.1. Chelation of heavy metals and sorption studies
102
M. Vandenbossche et al. / Journal of Environmental Management 143 (2014) 99e105
Finally, because the sorption process fits well the Langmuir model, the Gibbs free energy DG0 can also be calculated using equation (3) (Table 1). The same value is obtained for CuSO4 and Cu(NO3)2, proving that there is no influence of the counter-ion. The Gibbs free energy is more important for lead and above all for chromium: adsorption means strong interactions between the surface and the heavy metals. Thus, it can be concluded that heavy metals (i.e. copper (II), lead (II), and chromium (III)) sorption on PP-g-AA-cysteine is a spontaneous adsorption mechanism according to the thermodynamics of the system but that desorption is not spontaneous. To summarize, Cu(II), Pb(II) and Cr(III) adsorption on PP-g-AA-cysteine is favored energetically. Fig. 1. Amount of heavy metals trapped by the PP-g-AA-cysteine as a function of the equilibrium concentration in the contaminated solutions.
trapped (mg/g PP) as a function of the equilibrium concentration of heavy metals in the polluted solution. The textile functionalized with cysteine can chelate heavy metals increasingly with the equilibrium concentration of heavy metals in the solution until 500 mg/L where it reaches a plateau at about 21 mg/g PP for chromium (0.40 mmol/g PP), 95 mg/g PP for copper (1.5 mmol/g PP for SO2 4 counter-ion), 104 mg/g PP for copper (1.6 mmol/g PP for NO 3 counter-ion), and 135 mg/g for lead (0.66 mmol/g PP) (Fig. 1). Compared to PP-g-AA-cysteine, at 20 C for 24 h, in a 1000 mg/L heavy metal solution, virgin PP does not trap anything, and PP-g-AA adsorbs only 11 mg/g PP for chromium (0.21 mmol/g PP), 17 mg/g PP for copper (0.27 mmol/g PP for SO2 4 counter-ion), 18 mg/g PP for copper (0.28 mmol/g PP for NO 3 counter-ion), and 48 mg/g PP for lead (0.23 mmol/g PP). It thus shows that PP-g-AA can also chelate heavy metals but less efficiently than the cysteine grafted nonwoven. A previous study carried out with chitosan (Vandenbossche et al., 2013) gives lower sorption values: using a copper sulfate solution, only 30 mg copper per gram of PP were trapped, showing the efficiency of cysteine.
3.1.2. Sorption model At 1000 mg/L concentration, the textile is saturated with the four heavy metals used (whatever the counter-ion), and the curves obtained fit well the type I of the Brunauer classification (Brunauer et al., 1940). Thus Langmuir isotherm is the most adapted physical model for the heavy metals sorption on the textile surface. The equations obtained and the associated correlation coefficients (R2) are presented in Table 1. These results allow determining the theoretical maximum capacity of sorption qm and the Langmuir constant b (Table 1). This constant gives information on the adsorption and desorption mechanisms of heavy metals as b ¼ kads/ kdes. High values of b are obtained which means that the adsorption process is predominant over desorption. Moreover, chromium (b ¼ 40.57 L/mmol) is much more difficult to desorb than the other heavy metals tested.
3.1.3. EPR study To get further insight on the mechanism of heavy metals complexation, EPR-CW experiments were performed. Only the copper is considered in this part as Pb (II) state is EPR-silent, and as the amount of Cr (III) trapped is very small: the resolution was thus very low and it was not possible to determine the g values. The spectra given in Fig. 2 show typical copper d9 S ¼ 1/2 signal. For the PP-g-AA_Cu, a weak quasi only isotropic EPR signal centered at g ¼ 2.10 is observed. Such g value is in the range of g values for Cu2þ moiety (Peisach and Blumberg, 1974). When the spectrum related to the cysteine grafted material is considered, the EPR signal becomes anisotropic with two components: g// ¼ 2.61 and g⊥ ¼ 2.16. Thus, the coordination geometry is in an axial symmetry (Peisach and Blumberg, 1974). As copper has a nuclear spin I ¼ 3/2 (for both isotopes), a hyperfine interaction between electron and nuclear spin leads to a hyperfine constant of 142 G. According to the Peich Blumberg plot with the direct relation of g// versus A// for nature of atom that coordinate the Cu paramagnetic center can give information, a four oxygen coordination site is obtained, which indicates that the nitrogen of cysteine is not involved in the coordination (Peisach and Blumberg, 1974). Thus, copper is mainly coordinated with the carboxylic groups of cysteine, and also can be coordinated with the free carboxylic groups of acrylic acid. Moreover, the implication of sulfur cannot be observed in this case, mostly because of the small amount of sulfur at the surface of the fibers compared with oxygen (Vandenbossche et al., 2014).
Table 1 Data related to Langmuir isotherm in the trapping of copper, lead and chromium at 20 C and pH 4.5
Counter-ion R2 qm (mg/gPP) qm (mmol/gPP) b (L/mmol) DG0 (kJ/mol)
Cu (II)
Cu (II)
Pb (II)
Cr (III)
SO2 4 0.997 107 6 1.68 0.09 5.51 12.7
NO 3 0.998 117 7 1.8 0.1 5.71 12.6
NO 3 0.9994 147 10 0.71 0.04 10.07 11.2
NO 3 0.9995 21.3 0.3 0.410 0.006 40.57 7.8
Fig. 2. EPR spectra of PP-g-AA_Cu (a) and PP-g-AA-Cys_Cu (b).
M. Vandenbossche et al. / Journal of Environmental Management 143 (2014) 99e105
103
Fig. 3. Amount of heavy metals trapped versus time with the PP-g-AA-cysteine.
3.2. Kinetic studies Fig. 4. Amount of lead, chromium and copper trapped by the PP-g-AA-cysteine versus the pH of the contaminated solutions.
The evolution of the amount of heavy metals trapped by the cysteine grafted textile as a function of time has been determined for each artificially polluted solution at 20 C (Fig. 3). Heavy metals adsorption on PP-g-AA-cysteine is a short-time phenomenon, and the following order of adsorption capacity (expressed in mmol/g PP) is obtained: Cu(II) > Pb(II) > Cr(III). SRM models (Table 2) fit better the results than MTM models (Table S2). Thus, the sorption reaction order can then be determined considering the two kinetic models of adsorption. For each model, the rate constant of adsorption (k) and the correlation coefficient (R2) have been calculated (Table 2). The pseudo-second order kinetic model (Figure S2) fits better the sorption of copper and chromium whereas the pseudo-firstorder one (Figure S3) is better adapted to the sorption of lead. Thus the rate-controlling step for copper and chromium is the adsorption whereas it is the diffusion step in the case of lead. This result can be explained by the fact that copper and chromium are small atoms, and consequently, the diffusion of these atoms from the solution toward the surface is quicker than the diffusion of lead: lead needs bigger sites to be trapped, and thus needs more time to find these sites. The time corresponding to half the amount of heavy metals trapped (compared to the maximum amount that can be trapped) can then be calculated depending on the selected order (Table 2). The calculated t1/2 times vary between 5 min 40 s and 11 min 33 s, which means that the adsorption rate depends on the heavy metal used. Copper is trapped much more quickly than chromium and lead. The following trapping rate ranking is obtained: Cu(II) > Pb(II) w Cr(III). Moreover, the counter-ion used for copper does not show any influence on the trapping rate.
chromium and copper were identified: pH ¼ 6 and pH ¼ 5 respectively. In these optimized conditions, the amount of heavy metals trapped by the grafted nonwoven is around 28 mg Cr, 96 mg Cu (CuSO4), and 105 mg Cu (Cu(NO3)2) per gram of PP. The optimum pH value obtained for copper can be explained by the fact that, from around pH ¼ 5.0, the precipitation of Cu2þ in Cu(OH)2 decreases the copper concentration in the solution, and thus less Cu2þ can be trapped. When the pH is lower than 4.5, the decrease of the copper sorption can be explained by the continuous regeneration of the surface by the Hþ ions in the solution. Concerning the chromium sorption, when the pH is too low (pH 2.7), the same phenomenon occurs. However, when the pH increases, chromium is trapped increasingly until pH ¼ 6 and afterwards the adsorption capacity of the textile decreases. This phenomenon can be explained by the speciation of chromium (Fig. 5): a variety of chromium forms are present (Cr3þ, CrOH2þ, 5þ þ 3 4 Cr2(OH)4þ 2 , Cr (OH)4 , Cr(OH)2 , Cr(OH)3 and Cr(OH) ) (Dos Santos et al., 2012). It can be assumed that chromium cannot be trapped if too many hydroxyl groups are present: hydrogen bonds may be created but they do not withstand the washing step and heavy metals are then released. As for lead, from pH 2.7 to 8, the amount of metal trapped is continuously increasing with a maximum amount of 156 mg/g PP (0.75 mmol/g PP) at pH 8. This result is very interesting as the usual pH values of seawater and sediments are respectively around 8.2 and 7.8. The result obtained for lead proves that the adsorption capacity of the PP-g-AA-cysteine textile increases with pH. Indeed, when pH increases, the surface becomes more and more anionic, favoring the electronic attraction with heavy metal cations. Thus, an important amount of cationic heavy metals (from 150 to 156 mg Pb per gram PP) which do not precipitate between pH 7 and 8 can be trapped at the surface of the textile in this pH range.
3.3. Influence of pH Fig. 4 shows the evolution of the amount of heavy metals trapped as a function of the pH of the contaminated solution. pH has a significant effect on the sorption process (Fig. 3). Indeed, at 20 C, optimum pH corresponding to the maximum trapping of
Table 2 k1 (min1), k2 (g mmol1/min1), qe (mmol/g PP) and R2 values for PFO and PSO models, selected reaction order and corresponding t1/2 value for the sorption of heavy metals on the PP-g-AA-cysteine. Experimental data qe
CuSO4 Cu(NO3)2 Pb(NO3)2 Cr(NO3)3
1.5 1.6 0.66 0.40
Pseudo-first order
Pseudo-second order
k1
qe1
R2
k2
qe2
R2
0.086 0.087 0.059 0.066
1.46 1.62 0.67 0.37
0.991 0.996 0.96 0.94
0.087 0.086 0.115 0.282
1.56 1.71 0.73 0.38
0.9992 0.995 0.92 0.98
Selected reaction order
t1/2
PSO PFO/PSO PFO PSO
5 min 52 s 5 min 40 s 11 min 33 s 8 min 47 s
104
M. Vandenbossche et al. / Journal of Environmental Management 143 (2014) 99e105
Fig. 5. Chromium (III) speciation diagram [14].
Fig. 7. Amount of heavy metals trapped (mg/gPP) by the PP-g-AA-cysteine in the solution containing a mixture of heavy metals with and without NaCl.
3.4. Influence of the sodium concentration 3.5. Influence of a mixture of heavy metals in the solution Presence of sodium plays an important role in the heavy metals sorption. Fig. 6 shows that the higher the sodium concentration, the lower the amount of heavy metals trapped. Indeed, adsorption capacity of lead decreases quite linearly as a function of the NaCl concentration from 135 mg/g to 74 mg/g PP, i.e. a reduction of 45% when the textile is put into a heavy metals containing solution (1000 mg/L, pH ¼ 4.5, 20 C, 24 h) containing 30 g/L NaCl. The grafted textile is thus 1.8 times less efficient to decontaminate lead in this model solution of seawater. The tendency is similar for copper and chromium: the adsorption capacity of copper decreases from 95 to 47 mg/g PP and from 104 to 53 mg/g PP, for CuSO4 and Cu(NO3)2 solutions respectively, i.e. a reduction of 50%, and the adsorption capacity of chromium decreases from 21 to 6 mg/g PP in a solution containing 30 g/L NaCl, corresponding to a reduction of 71%. Thus, the grafted textile is twice less efficient to decontaminate copper, and 3.5 times less efficient to decontaminate chromium. This phenomenon can be explained by a competition between sodium and the heavy metals at the textile surface (Liu et al., 2013). PP-g-AA-cysteine can trap heavy metals in solutions containing copper, chromium or lead with the same counter-ion (NO-3). But, in leachates or sediments, there is rather a mixture of heavy metals. Thus, sorption studies have been carried out in a solution containing a mixture of heavy metals (copper, lead and chromium), in order to determine the behavior of the samples when there is a competition between several metals.
The competition between heavy metals to enter the reactive sites at the surface of the functionalized textile plays an important role in the adsorption efficiency of the material. Indeed, the competition between copper, chromium and lead is due to both the limited sites number and the size of the heavy metals. Sorption kinetics, as well as pH and ionic strength, could have an impact on the amount of heavy metals trapped. Fig. 7 shows that in a solution containing a mixture of heavy metals, the amount of ions trapped at the surface of the sample for each heavy metal is inferior to the amount trapped with the solution containing only the considered heavy metal. A maximum of copper can be trapped at pH 4.5 (33 mg Cu/g PP in the solution without sodium), and a maximum of chromium is trapped at pH 6.0 (13 mg Cr/g PP in the solution without sodium). At higher pH, the chromium concentration on PPg-AA-cysteine is too low to be detected by the atomic absorption spectrometer. At high pH (higher than 6), copper and chromium precipitate and thus the concentration of these heavy metals in the solution decreases. Then, less copper and chromium can be trapped by the functionalized textile and thus there are more and more free sites at the surface of the PP-g-AA-cysteine to trap lead. A maximum sorption capacity is observed at pH 8.0, with 63 mg Pb/g PP without sodium, which is much less important than the results obtained for the sorption of lead in a single metal containing solution. The presence of sodium also decreases the sorption efficiency of the PP-g-AA-cysteine. As observed for the single metal containing solutions, the amounts of copper and lead trapped decrease, due to the competition with sodium at the surface of PP-g-AA-cysteine. However, amounts of chromium trapped at pH 4.5 and 6.0 being similar whatever the sodium concentration, chromium adsorption seems not to be affected by the presence of sodium, but much more by the presence of copper and lead. This result could be explained by the small amount of chromium trapped. In the case of copper and lead, presence of sodium has an important effect on the sorption capacity: at pH 4.5, only 13 mg Cu/g PP are trapped instead of 33 mg Cu/g PP without sodium. 4. Conclusion
Fig. 6. Amount of lead, chromium and copper trapped by the PP-g-AA-cysteine as a function of the NaCl concentration in the contaminated solutions.
PP-g-AA-cysteine is a promising material in heavy metals depollution. Indeed, Cu(II), Pb(II) and Cr(III) can be trapped in interesting amounts from aqueous media using this material. Copper is the most interesting heavy metal because of the high
M. Vandenbossche et al. / Journal of Environmental Management 143 (2014) 99e105
amount that can be trapped and because of the adsorption kinetics: copper is trapped much quicker than the other heavy metals. EPR studies give information on the reactive functions involved in the metal coordination, and more precisely in the case of copper. Using this technique, it was shown that carboxylic acid groups, which are the most abundant chemical functions at the surface of the fibers, are involved in the trapping of copper. Thus, as these chemical groups can easily interact with sodium, it can explain the important decrease of the sorption capacity in the presence of NaCl: 45% for lead, 50% for copper and 71% for chromium. It was also observed that the adsorption capacity of the textile increases at high pH values: the material has an anionic structure at high pH, and thus the attraction between the anionic surface and the cationic heavy metals increases. But, generally, heavy metals precipitate at high pH (from pH 5e6), which does not allow the sorption of these heavy metals. This result is interesting as pH of marine sediments is about 8, thus this new material can be efficient to trap heavy metals at this pH. Acknowledgments The authors gratefully acknowledge the support of FEDER (Fonds Européen de Développement Régional), Nord-Pas-de-Calais region, and FUI (Fonds Unique Interministériel) for funding this work .We also would like to deeply acknowledge the support of the Up-Tex Competitiveness Cluster and all DEPOLTEX partners for helpful collaboration and discussion. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.jenvman.2014.05.002. References Brunauer, S., Deming, L.S., Deming, W.E., Teller, E., 1940. On a theory of the van der Waals adsorption of gases. J. Am. Chem. Soc. 62, 1723e1732. Chatterjee, A., Schiewer, S., 2014. Multi-resistance kinetic models for biosorption of Cd by raw and immobilized citrus peels in batch and packed-bed columns. Chem. Eng. J. 244, 105e116.
105
Chiou, M.S., Li, H.Y., 2003. Adsorption behavior of reactive dye in aqueous solution on chemical cross-linked chitosan beads. Chemosphere 50, 1095e1105. Chu, K.H., 2002. Removal of copper from aqueous solution by chitosan in prawn shell: adsorption equilibrium and kinetics. J. Hazard. Mater. 90, 77e95. Cobbett, C.S., 2000. Phytochelatins and their roles in heavy metal detoxification. Plant Physiol. 123, 825e832. Cooney, D.O., 1998. Adsorption Design for Wastewater Treatment. CRC Press. Dos Santos, V.C.G., Salvado, A.d.P.A., Dragunski, D.C., Peraro, D.N.C., Tarley, C.R.T., Caetano, J., 2012. Highly improved chromium (III) uptake capacity in modified sugarcane bagasse using different chemical treatments. Quim. Nova 35, 1606e 1611. Eggleton, J., Thomas, K.V., 2004. A review of factors affecting the release and bioavailability of contaminants during sediment disturbance events. Environ. Int. 30, 973e980. Foo, K.Y., Hameed, B.H., 2010. Insights into the modeling of adsorption isotherm systems. Chem. Eng. J. 156, 2e10. Gerente, C., Lee, V.K.C., Le Cloirec, P., McKay, G., 2007. Application of chitosan for the removal of metals from wastewaters by adsorption-mechanisms and models review. Crit. Rev. Env. Sci. Tec. 37, 41e127. Gouzy, A., Ducos, G., 2008. La connaissance des éléments-traces métalliques: un défi pour la gestion de l’environnement. Air Pur 75, 6e10. Klaassen, C.D., Liu, J., Choudhuri, S., 1999. METALLOTHIONEIN: an intracellular protein to protect against cadmium toxicity. Annu. Rev. Pharmacol. Toxicol. 39, 267e294. Langmuir, I., 1918. The adsorption of gases on plane surfaces of glass, mica and platinum. J. Am. Chem. Soc. 40, 1361e1403. Liu, W., Wang, T., Borthwick, A.G.L., Wang, Y., Yin, X., Li, X., Ni, J., 2013. Adsorption of Pb2þ, Cd2þ, Cu2þ and Cr3þ onto titanate nanotubes: competition and effect of inorganic ions. Sci. Total Environ. 456e457, 171e180. Maret, W., Larsen, K.S., Vallee, B.L., 1997. Coordination dynamics of biological zinc “clusters” in metallothioneins and in the DNA-binding domain of the transcription factor Gal4. Proc. Natl. Acad. Sci. 94, 2233e2237. Peisach, J., Blumberg, W.E., 1974. Structural implications derived from the analysis of electron paramagnetic resonance spectra of natural and artificial copper proteins. Arch. Biochem. Biophys. 165, 691e708. 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 Univ. Sci. A 10, 716e724. Singh, S.P., Tack, F.M.G., Verloo, M.G., 1998. Land disposal of heavy metal contaminated dredged sediments: a review of environmental aspects. Land Contam. Recl. 6, 149e158. Stronkhorst, J., Schot, M.E., Dubbeldam, M.C., Ho, K.T., 2003. A toxicity identification evaluation of silty marine harbor sediments to characterize persistent and nonpersistent constituents. Mar. Pollut. Bull. 46, 56e64. Vandenbossche, M., Casetta, M., Jimenez, M., Bellayer, S., Traisnel, M., 2014. Cysteine-grafted nonwoven geotextile: a new and efficient material for heavy metals sorption e part A. J. Environ. Manage 132, 107e112. Vandenbossche, M., Jimenez, M., Casetta, M., Bellayer, S., Beaurain, A., Bourbigot, S., Traisnel, M., 2013. Chitosan-grafted nonwoven geotextile for heavy metals sorption in sediments. React. Funct. Polym. 73, 53e59. Zenk, M.H., 1996. Heavy metal detoxification in higher plants e a review. Gene 179, 21e30.
Contents lists available at ScienceDirect
Journal of Environmental Management journal homepage: www.elsevier.com/locate/jenvman
Cysteine-grafted nonwoven geotextile: A new and efficient material for heavy metals sorption e Part B M. Vandenbossche a, H. Vezin b, N. Touati b, M. Jimenez a, M. Casetta a, *, M. Traisnel a a Unité Matériaux et Transformations (UMET), Ingénierie des Systèmes Polymères (ISP), CNRS-UMR 8207, ENSCL, Université Lille Nord de France, 59652 Villeneuve d’Ascq Cedex, France b Laboratoire de Spectrochimie Infrarouge et Raman (LASIR), UMR-CNRS 8516, Université Lille Nord de France, Bâtiment C5, 59655 Villeneuve d’Ascq Cedex, France
a r t i c l e i n f o
a b s t r a c t
Article history: Received 14 January 2014 Received in revised form 1 April 2014 Accepted 2 May 2014 Available online
The development of a new material designed to trap heavy metals from sediments or wastewater, based on a polypropylene non-woven covalently grafted with cysteine, has been reported in a previous paper (Part A). The non-woven was first functionalized with acrylic acid (AA) which is used as spacer, and then cysteine was immobilized on the substrate through covalent coupling in order to obtain the so-called PP-g-AA-cysteine. Some preliminary heavy metals adsorption tests gave interesting results: at 20 C for 24 h and in a 1000 mg/L heavy metals solution, PP-g-AA-cysteine adsorbs 95 mg Cu/g PP (CuSO4 solution), 104 mg Cu/g PP (Cu(NO3)2 solution), 135 mg Pb/g PP (Pb(NO3)2 solution) and 21 mg Cr/g PP (Cr(NO3)3 solution). In this second part of the work, heavy metals sorption tests were carried out with Cu (II), Pb (II), and Cr (III) separately, in order to determine the sorption capacity of this new sorbent as a function of (i) the heavy metals concentration in the solution, (ii) the contact time with the solution, (iii) the pH and (iv) the ionic strength of the solution containing heavy metals. Moreover, the sorption capacity of PP-g-AA-Cysteine was studied using a polluted solution consisting of a mixture of these different heavy metals. An Electron Paramagnetic Resonance study was finally carried out in order to determine the coordination geometry in the environment of the copper trapped by the PP-g-AAcysteine. Ó 2014 Elsevier Ltd. All rights reserved.
Keywords: Adsorption Heavy metals EPR Polypropylene Remediation
1. Introduction Currently, the accumulation of massive amounts of sediments leads to important economic and environmental issues. In order to maintain the depth of the navigational waterways, harbors and estuaries worldwide (Stronkhorst et al., 2003), and also to limit the remobilization of contaminants which are present in sediments (Eggleton and Thomas, 2004), dredging operations occur regularly. But most of these dredged sediments can be considered as waste as they contain harmful components such as organic compounds and heavy metals (Singh et al., 1998). Because of the high concentration in heavy metals such as Arsenic, Cadmium, Chromium, Copper, Mercury, Nickel, Lead, and Zinc (Gouzy and Ducos, 2008), these sediments must be put in storage centers. But as storage centers are
* Corresponding author. E-mail address: [email protected] (M. Casetta). http://dx.doi.org/10.1016/j.jenvman.2014.05.002 0301-4797/Ó 2014 Elsevier Ltd. All rights reserved.
saturated, sediments have to be treated in order to remove heavy metals and to give them a second life. Geotextiles are commonly used in filtration and drainage. The grafting of specific biomolecules provides heavy metals adsorption properties to the textile. Cysteine was chosen as a great number of chelating biomolecules (such as glutathione, phytochelatins and metallothioneins) contain a large amount of cysteine units (Cobbett, 2000; Klaassen et al., 1999; Maret et al., 1997; Zenk, 1996). The grafting of cysteine on polypropylene nonwovens (PP) was already carried out and optimized in our group and is detailed in the Part A of this study (Vandenbossche et al., 2014). In the present paper, a heavy metals sorption investigation is proposed, in order to determine the capacity of this new material, PP-g-AA-cysteine for short, to trap copper (II), lead (II), and chromium (III) in various experimental conditions, i.e. when initial metal concentration, contact time, pH and ionic strength vary. Electron Paramagnetic Resonance (EPR) analyses were also carried out on copper solutions and on PP-g-AA-cysteine e copper in order to determine the heteroatoms involved in the copper sorption.
100
M. Vandenbossche et al. / Journal of Environmental Management 143 (2014) 99e105
2. Materials and methods
Thermo Scientific and Mono-elements coded hollow-cathodes lamp for Pb, Thermo Scientific).
2.1. Preparation of the optimized PP-g-AA-cysteine Squares (5*5 cm2) were cut from polypropylene (PP) nonwoven INTN50 (50 g/m2, provided by PGI nonwovens, France), washed and dried as already described in a previous paper (Vandenbossche et al., 2013). Acrylic acid (purity 99.5%, provided by Acros Organics) was graft-polymerized on PP by a cold plasma process (PPg-AA) also previously optimized and reported (Vandenbossche et al. 2013). Then cysteine (purity 97%, provided by Sigmae Aldrich) was immobilized on PP-g-AA by a chemical coupling between the carboxylic acid group of acrylic acid and the amine group of cysteine using a carbodiimide: N-(3-dimethylaminopropyl)-N0 ethylcarbodiimide hydrochloride (purity 98%, provided by SigmaeAldrich). The immobilization of cysteine was carried out according to the optimized conditions determined in the Part A (Vandenbossche et al., 2014): PP-g-AA was first immersed in a solution containing the carbodiimide (1 h at 4 C), and then immersed in a 0.229 mol/L cysteine solution for 28 h at 20 C. Samples were then washed for 1 h in an ultrasonic bath, and for 6 h in a Soxhlet extractor with distilled water. Finally, PP-g-AA-cysteine samples were dried under vacuum for 16 h.
2.3.2. Preparation of the solution for the calibration of the atomic absorption spectrometer The determination of the concentration of heavy metals in the solution obtained after the digestion of the metals-chelated-textile was carried out by the standard addition method: the analysis was first made with a blank, and then with the solution to analyze of known dilution, and then with the same solution in which 1, 2, or 3 mg/L of heavy metals were added. 2.3.3. Influence of the metal concentration Cysteine-grafted-polypropylene samples were immersed in 100 mL of 100 mg/L, 250 mg/L, 500 mg/L, 1000 mg/L, and 5000 mg/L contaminated solutions at 20 C and pH ¼ 4.5 for 24 h. 2.3.4. Kinetic studies Cysteine-grafted polypropylene samples were immersed in 100 mL of 1000 mg/L contaminated solutions (Cu II, Pb II, Cr III) at 20 C and pH 4.5. For each heavy metal, the amount of metal trapped by the functionalized textile was determined as a function of time. It was then possible to determine the kinetic law of adsorption followed by the grafted surface.
2.2. Preparation of the artificially contaminated solutions Some artificially polluted solutions were prepared in order to determine the adsorption behavior of the PP-g-AA-cysteine as a function of the oxidation state, the ionic radius of the metal and the counter-ion. 1000 mg/L solutions of copper sulfate, copper nitrate, lead nitrate and chromium nitrate were prepared in a 1 L-volumetric flask, and the 5000 mg/L solutions were prepared in a 100 mL-volumetric flask. From the solutions containing 1000 mg/L heavy metals, various contaminated solutions were prepared, with the following concentrations: 100 mg/L, 250 mg/L, and 500 mg/L. In order to prepare the solutions containing 1000 mg/L and 5000 mg/L of heavy metals, copper sulfate solution was diluted in distilled water (conductivity ¼ 0.2 mS/cm) while copper nitrate, lead nitrate and chromium nitrate were dissolved (see supporting data, table S1) using these initial reagents: 0.1 M CuSO4.5H2O (provided by Sigma Aldrich), Cu(NO3)2.xH2O (purity ¼ 99.999%, provided by SigmaeAldrich), Pb(NO3)2 (purity 99.0%, provided by SigmaeAldrich), Cr(NO3)3 (purity ¼ 99.0%, provided by Sigmae Aldrich). Finally, a solution containing 1000 mg/L copper, 1000 mg/L chromium and 1000 mg/L lead was prepared by dissolving respectively 2.951 g of Cu(NO3)2, 7.700 g of Cr(NO3)3 and 1.600 g of Pb(NO3)2 in distilled water in a 1 L-volumetric flask. 2.3. Sorption studies 2.3.1. Chelation of heavy metals and sorption studies One cysteine-grafted polypropylene sample was immersed for 24 h in 100 mL of the previously described heavy metal solution thermostated at 20 C. The sample was then washed in a bath of 50 mL of ultrapure water (conductivity ¼ 0.055 mS/cm) in order to remove heavy metals located at the surface which did not interact with cysteine or acrylic acid. Then the textile was digested with sulfuric acid at 95% and hydrochloric acid at 37% (provided by VWR BDH Prolabo) to remove heavy metals which interacted with the surface. Ultrapure water was added to the solution, and the concentration of the diluted heavy metal containing solution was determined using flame atomic absorption spectrometry (Thermo Solaar S4 AA Spectrometer, Thermo S Series, Multi-elements combined coded hollow-cathode lamp for CreCueMneNi,
2.3.5. Influence of pH Cysteine-grafted-polypropylene samples were immersed for 24 h in 100 mL of 1000 mg/L contaminated solutions at 20 C. 1M HCl or 1M NaOH was added in order to reach the following pH: 2.7, 4.5, 6.0, or 8.0. 2.3.6. Influence of the sodium concentration Cysteine-grafted-polypropylene samples were immersed in 100 mL of 1000 mg/L contaminated solutions in which NaCl (Sodium chloride for analysis EMSUREÒ ACS, ISO, Reag. Ph Eur, Merck) was added to reach the following concentrations: 10, 20, or 30 g/L. These experiments were carried out at 20 C and pH ¼ 4.5. 2.3.7. Influence of a mixture of heavy metals in the solution Cysteine-grafted-polypropylene samples were immersed in 100 mL of an artificially contaminated solution containing 1000 mg/L of copper, 1000 mg/L of lead and 1000 mg/L of chromium (for all these heavy metals the counter-ion was NO 3 ) at pH ¼ 4.5, 6, 7 and 8 in presence of 30 g/L NaCl or without NaCl. 2.4. Physical characterizations of the grafted textile 2.4.1. Langmuir isotherm According to the classification of Brunauer et al. (1940), the Langmuir isotherm is the first physical model of sorption (type I). It is an empirical sorption model (Foo and Hameed, 2010) which characterizes the adsorption of a monolayer of compounds on a microporous material. In this model, (i) the energy is assumed to be the same all over the surface, (ii) the trapped molecules are assumed not to interact together or migrate over the surface and (iii) at the maximum of sorption, only one monolayer can be formed on the free surface. In the case of the adsorption on a solid surface of a molecule from a solution, the Langmuir isotherm model can be expressed according to the following equation (Langmuir, 1918):
q ¼ qm *b*C=ð1 þ b*CÞ
(1)
where q (mg/g PP) is the sorption capacity, qm (mg/g PP) is the maximum sorption capacity, C (mol/L) is the concentration of heavy metal in the solution, and b (L/mol) is the Langmuir constant. In this
M. Vandenbossche et al. / Journal of Environmental Management 143 (2014) 99e105
101
case, the free Gibbs energy can be calculated with the following formula:
homogeneous sphere (Cooney, 1998). The HSDM equation has the following form:
DG0 ¼ RTlnðbÞ
. . . vq=vt ¼ DS r2 v r2 vq vr vr
(2)
where R (8.31 J/mol/K) is the perfect gas constant, and T (K) is the temperature.
2.4.2. Kinetic studies: pseudo-first-order and pseudo-second-order The adsorption mechanism is assumed to occur in three steps (Chiou and Li, 2003): first, the molecule moves through the solution, then from the solution towards the surface which corresponds to the diffusional step that can be divided into two stages (1) diffusion across the liquid film surrounding the textile (external diffusion or film diffusion) and (2) diffusion in the liquid contained in the pores and/or along the pore walls (internal diffusion or intraparticle diffusion) (Qiu et al., 2009); and in a third step, the molecule is trapped at the surface of the solid material which corresponds to the adsorption onto active sites. Literature listed a lot of kinetic laws allowing the evaluation of the adsorbents performance and giving an idea concerning the implicit mechanism (Chatterjee and Schiewer, 2014; Chu, 2002). In this paper, some kinetic laws were chosen because of their efficiency to describe adsorption of compounds from aqueous media onto solid surfaces: the intraparticle diffusion models, homogeneous solid diffusion model (HSDM) and linear driving force (LDF) (Qiu et al., 2009) which are mass transfer modeling (MTM) and also the two well-known kinetic models used in this field the pseudo-first-order (PFO) and pseudo-second-order (PSO) equations, which are surface reaction modeling (SRM) (Gerente et al., 2007). As it was reported by Chatterjee and Schiewer, the intrinsic sorption reaction is assumed to be fast in MTM, and thus, the diffusion processes are considered as rate limiting steps, whereas in SRM, the mass transfer is assumed to be fast and the trapping of the adsorbent is the rate limiting step (Chatterjee and Schiewer, 2014). The first equation of the sorption capacity in a solid/liquid system, for example the sorption of heavy metals in solution on a material surface, is called the Lagergren rate equation or pseudo-first-order. This kinetic model has the following form (Gerente et al., 2007):
dqt =dt ¼ k1 ðqe qt Þ
(3)
where qe and qt (mg/g) are respectively the sorption capacities at equilibrium and at time t, and k1 is the pseudo-first-order sorption rate constant. The pseudo-first-order equation in linear form is the following one:
lnðqe qt Þ ¼ lnðqe Þ k1 t
(4)
However, the pseudo-first-order equation cannot explain all the kinetic phenomena, and it is the reason why other models have been developed. Generally, the pseudo-second-order model is much more successful, and has the following form (Gerente et al., 2007):
dqt =dt ¼ k2 ðqe qt Þ2
(5)
(7)
where DS is the intraparticle diffusion coefficient, r the average particle radius, q the adsorption capacity and t the time. An exact solution can be given for the “infinite bath” case where the system is considered free of adsorbate and the adsorbate concentration at the surface remains constant (Qiu et al., 2009). Some approximations can be carried out in order to simplify the equation. Indeed, particles are considered spherical at any particular time and the external film resistance can be neglected according to the constant surface concentration (Cooney, 1998). Thus, two solutions are considered depending on the sorption time. For a short time (when qa/qN < 0.3 with qa the average value of q, and qN the average concentration in the solid at infinite time), the equation can be simplified as:
1=2 . qa =qN ¼ 6* DS R2 *p *t 1=2
(8)
where R is the total particle radius. For a long time, the equation is simplified as:
. . qa =qN ¼ 1 6 p2 exp DS *p2 *t R2
(9)
Equation (8) assumes that adsorption rate decreases along with the increasing adsorbent particle size and vice versa (Qiu et al., 2009). Concerning Equation (9), the assumption of constant surface concentration for HSDM is not likely to be verified at a long time, and consequently, only short time experiments are considered (Cooney, 1998). Intraparticle diffusion can also be modeled using the linear driving force (LDF) approximation. This model assumes that the rate of mass transfer linearly depends on the concentration difference between the external surface of the adsorbent and the solution. The following equation corresponds to this model:
dq=dt ¼ kS *S0 *ðqi qÞ
(10)
where S0 is the particle surface area per particle volume and kS the transfer coefficient. 2.4.3. Electronic paramagnetic resonance spectroscopy EPR spectroscopy experiments were performed at X band using a Brüker ELEXSYS E500 spectrometer in continuous waves (CW). The spectra were recorded on TM cavity with 10 mW of microwave power and with modulation amplitude of 4 G. The spectra were collected at room temperature on the grafted polypropylene directly inserted in a 8 mm quartz tube. The g-factor, which is a constant of proportionality, was determined in order to have information on the properties of the electron in a certain environment. When there is a hyperfine structure, it is possible to determine the hyperfine constant A which corresponds to the distance between two consecutive stripes.
where qe and qt (mg/g) are the sorption capacities at equilibrium and at time t respectively, and k2 is the pseudo-second-order sorption rate constant. The pseudo-second-order equation in linear form is as follows:
3. Results and discussion
1=ðqe qt Þ ¼ k2 t þ 1=qe
3.1.1. Sorption capacity of the material The sorption capacity of the textile depends on the heavy metals concentration of the artificially contaminated solution. Fig. 1 (and Figure S1) shows the evolution of the amount of heavy metals
(6)
The homogeneous solid diffusion model (HSDM) is a typical intraparticle diffusion model which can describe mass transfer in a
3.1. Chelation of heavy metals and sorption studies
102
M. Vandenbossche et al. / Journal of Environmental Management 143 (2014) 99e105
Finally, because the sorption process fits well the Langmuir model, the Gibbs free energy DG0 can also be calculated using equation (3) (Table 1). The same value is obtained for CuSO4 and Cu(NO3)2, proving that there is no influence of the counter-ion. The Gibbs free energy is more important for lead and above all for chromium: adsorption means strong interactions between the surface and the heavy metals. Thus, it can be concluded that heavy metals (i.e. copper (II), lead (II), and chromium (III)) sorption on PP-g-AA-cysteine is a spontaneous adsorption mechanism according to the thermodynamics of the system but that desorption is not spontaneous. To summarize, Cu(II), Pb(II) and Cr(III) adsorption on PP-g-AA-cysteine is favored energetically. Fig. 1. Amount of heavy metals trapped by the PP-g-AA-cysteine as a function of the equilibrium concentration in the contaminated solutions.
trapped (mg/g PP) as a function of the equilibrium concentration of heavy metals in the polluted solution. The textile functionalized with cysteine can chelate heavy metals increasingly with the equilibrium concentration of heavy metals in the solution until 500 mg/L where it reaches a plateau at about 21 mg/g PP for chromium (0.40 mmol/g PP), 95 mg/g PP for copper (1.5 mmol/g PP for SO2 4 counter-ion), 104 mg/g PP for copper (1.6 mmol/g PP for NO 3 counter-ion), and 135 mg/g for lead (0.66 mmol/g PP) (Fig. 1). Compared to PP-g-AA-cysteine, at 20 C for 24 h, in a 1000 mg/L heavy metal solution, virgin PP does not trap anything, and PP-g-AA adsorbs only 11 mg/g PP for chromium (0.21 mmol/g PP), 17 mg/g PP for copper (0.27 mmol/g PP for SO2 4 counter-ion), 18 mg/g PP for copper (0.28 mmol/g PP for NO 3 counter-ion), and 48 mg/g PP for lead (0.23 mmol/g PP). It thus shows that PP-g-AA can also chelate heavy metals but less efficiently than the cysteine grafted nonwoven. A previous study carried out with chitosan (Vandenbossche et al., 2013) gives lower sorption values: using a copper sulfate solution, only 30 mg copper per gram of PP were trapped, showing the efficiency of cysteine.
3.1.2. Sorption model At 1000 mg/L concentration, the textile is saturated with the four heavy metals used (whatever the counter-ion), and the curves obtained fit well the type I of the Brunauer classification (Brunauer et al., 1940). Thus Langmuir isotherm is the most adapted physical model for the heavy metals sorption on the textile surface. The equations obtained and the associated correlation coefficients (R2) are presented in Table 1. These results allow determining the theoretical maximum capacity of sorption qm and the Langmuir constant b (Table 1). This constant gives information on the adsorption and desorption mechanisms of heavy metals as b ¼ kads/ kdes. High values of b are obtained which means that the adsorption process is predominant over desorption. Moreover, chromium (b ¼ 40.57 L/mmol) is much more difficult to desorb than the other heavy metals tested.
3.1.3. EPR study To get further insight on the mechanism of heavy metals complexation, EPR-CW experiments were performed. Only the copper is considered in this part as Pb (II) state is EPR-silent, and as the amount of Cr (III) trapped is very small: the resolution was thus very low and it was not possible to determine the g values. The spectra given in Fig. 2 show typical copper d9 S ¼ 1/2 signal. For the PP-g-AA_Cu, a weak quasi only isotropic EPR signal centered at g ¼ 2.10 is observed. Such g value is in the range of g values for Cu2þ moiety (Peisach and Blumberg, 1974). When the spectrum related to the cysteine grafted material is considered, the EPR signal becomes anisotropic with two components: g// ¼ 2.61 and g⊥ ¼ 2.16. Thus, the coordination geometry is in an axial symmetry (Peisach and Blumberg, 1974). As copper has a nuclear spin I ¼ 3/2 (for both isotopes), a hyperfine interaction between electron and nuclear spin leads to a hyperfine constant of 142 G. According to the Peich Blumberg plot with the direct relation of g// versus A// for nature of atom that coordinate the Cu paramagnetic center can give information, a four oxygen coordination site is obtained, which indicates that the nitrogen of cysteine is not involved in the coordination (Peisach and Blumberg, 1974). Thus, copper is mainly coordinated with the carboxylic groups of cysteine, and also can be coordinated with the free carboxylic groups of acrylic acid. Moreover, the implication of sulfur cannot be observed in this case, mostly because of the small amount of sulfur at the surface of the fibers compared with oxygen (Vandenbossche et al., 2014).
Table 1 Data related to Langmuir isotherm in the trapping of copper, lead and chromium at 20 C and pH 4.5
Counter-ion R2 qm (mg/gPP) qm (mmol/gPP) b (L/mmol) DG0 (kJ/mol)
Cu (II)
Cu (II)
Pb (II)
Cr (III)
SO2 4 0.997 107 6 1.68 0.09 5.51 12.7
NO 3 0.998 117 7 1.8 0.1 5.71 12.6
NO 3 0.9994 147 10 0.71 0.04 10.07 11.2
NO 3 0.9995 21.3 0.3 0.410 0.006 40.57 7.8
Fig. 2. EPR spectra of PP-g-AA_Cu (a) and PP-g-AA-Cys_Cu (b).
M. Vandenbossche et al. / Journal of Environmental Management 143 (2014) 99e105
103
Fig. 3. Amount of heavy metals trapped versus time with the PP-g-AA-cysteine.
3.2. Kinetic studies Fig. 4. Amount of lead, chromium and copper trapped by the PP-g-AA-cysteine versus the pH of the contaminated solutions.
The evolution of the amount of heavy metals trapped by the cysteine grafted textile as a function of time has been determined for each artificially polluted solution at 20 C (Fig. 3). Heavy metals adsorption on PP-g-AA-cysteine is a short-time phenomenon, and the following order of adsorption capacity (expressed in mmol/g PP) is obtained: Cu(II) > Pb(II) > Cr(III). SRM models (Table 2) fit better the results than MTM models (Table S2). Thus, the sorption reaction order can then be determined considering the two kinetic models of adsorption. For each model, the rate constant of adsorption (k) and the correlation coefficient (R2) have been calculated (Table 2). The pseudo-second order kinetic model (Figure S2) fits better the sorption of copper and chromium whereas the pseudo-firstorder one (Figure S3) is better adapted to the sorption of lead. Thus the rate-controlling step for copper and chromium is the adsorption whereas it is the diffusion step in the case of lead. This result can be explained by the fact that copper and chromium are small atoms, and consequently, the diffusion of these atoms from the solution toward the surface is quicker than the diffusion of lead: lead needs bigger sites to be trapped, and thus needs more time to find these sites. The time corresponding to half the amount of heavy metals trapped (compared to the maximum amount that can be trapped) can then be calculated depending on the selected order (Table 2). The calculated t1/2 times vary between 5 min 40 s and 11 min 33 s, which means that the adsorption rate depends on the heavy metal used. Copper is trapped much more quickly than chromium and lead. The following trapping rate ranking is obtained: Cu(II) > Pb(II) w Cr(III). Moreover, the counter-ion used for copper does not show any influence on the trapping rate.
chromium and copper were identified: pH ¼ 6 and pH ¼ 5 respectively. In these optimized conditions, the amount of heavy metals trapped by the grafted nonwoven is around 28 mg Cr, 96 mg Cu (CuSO4), and 105 mg Cu (Cu(NO3)2) per gram of PP. The optimum pH value obtained for copper can be explained by the fact that, from around pH ¼ 5.0, the precipitation of Cu2þ in Cu(OH)2 decreases the copper concentration in the solution, and thus less Cu2þ can be trapped. When the pH is lower than 4.5, the decrease of the copper sorption can be explained by the continuous regeneration of the surface by the Hþ ions in the solution. Concerning the chromium sorption, when the pH is too low (pH 2.7), the same phenomenon occurs. However, when the pH increases, chromium is trapped increasingly until pH ¼ 6 and afterwards the adsorption capacity of the textile decreases. This phenomenon can be explained by the speciation of chromium (Fig. 5): a variety of chromium forms are present (Cr3þ, CrOH2þ, 5þ þ 3 4 Cr2(OH)4þ 2 , Cr (OH)4 , Cr(OH)2 , Cr(OH)3 and Cr(OH) ) (Dos Santos et al., 2012). It can be assumed that chromium cannot be trapped if too many hydroxyl groups are present: hydrogen bonds may be created but they do not withstand the washing step and heavy metals are then released. As for lead, from pH 2.7 to 8, the amount of metal trapped is continuously increasing with a maximum amount of 156 mg/g PP (0.75 mmol/g PP) at pH 8. This result is very interesting as the usual pH values of seawater and sediments are respectively around 8.2 and 7.8. The result obtained for lead proves that the adsorption capacity of the PP-g-AA-cysteine textile increases with pH. Indeed, when pH increases, the surface becomes more and more anionic, favoring the electronic attraction with heavy metal cations. Thus, an important amount of cationic heavy metals (from 150 to 156 mg Pb per gram PP) which do not precipitate between pH 7 and 8 can be trapped at the surface of the textile in this pH range.
3.3. Influence of pH Fig. 4 shows the evolution of the amount of heavy metals trapped as a function of the pH of the contaminated solution. pH has a significant effect on the sorption process (Fig. 3). Indeed, at 20 C, optimum pH corresponding to the maximum trapping of
Table 2 k1 (min1), k2 (g mmol1/min1), qe (mmol/g PP) and R2 values for PFO and PSO models, selected reaction order and corresponding t1/2 value for the sorption of heavy metals on the PP-g-AA-cysteine. Experimental data qe
CuSO4 Cu(NO3)2 Pb(NO3)2 Cr(NO3)3
1.5 1.6 0.66 0.40
Pseudo-first order
Pseudo-second order
k1
qe1
R2
k2
qe2
R2
0.086 0.087 0.059 0.066
1.46 1.62 0.67 0.37
0.991 0.996 0.96 0.94
0.087 0.086 0.115 0.282
1.56 1.71 0.73 0.38
0.9992 0.995 0.92 0.98
Selected reaction order
t1/2
PSO PFO/PSO PFO PSO
5 min 52 s 5 min 40 s 11 min 33 s 8 min 47 s
104
M. Vandenbossche et al. / Journal of Environmental Management 143 (2014) 99e105
Fig. 5. Chromium (III) speciation diagram [14].
Fig. 7. Amount of heavy metals trapped (mg/gPP) by the PP-g-AA-cysteine in the solution containing a mixture of heavy metals with and without NaCl.
3.4. Influence of the sodium concentration 3.5. Influence of a mixture of heavy metals in the solution Presence of sodium plays an important role in the heavy metals sorption. Fig. 6 shows that the higher the sodium concentration, the lower the amount of heavy metals trapped. Indeed, adsorption capacity of lead decreases quite linearly as a function of the NaCl concentration from 135 mg/g to 74 mg/g PP, i.e. a reduction of 45% when the textile is put into a heavy metals containing solution (1000 mg/L, pH ¼ 4.5, 20 C, 24 h) containing 30 g/L NaCl. The grafted textile is thus 1.8 times less efficient to decontaminate lead in this model solution of seawater. The tendency is similar for copper and chromium: the adsorption capacity of copper decreases from 95 to 47 mg/g PP and from 104 to 53 mg/g PP, for CuSO4 and Cu(NO3)2 solutions respectively, i.e. a reduction of 50%, and the adsorption capacity of chromium decreases from 21 to 6 mg/g PP in a solution containing 30 g/L NaCl, corresponding to a reduction of 71%. Thus, the grafted textile is twice less efficient to decontaminate copper, and 3.5 times less efficient to decontaminate chromium. This phenomenon can be explained by a competition between sodium and the heavy metals at the textile surface (Liu et al., 2013). PP-g-AA-cysteine can trap heavy metals in solutions containing copper, chromium or lead with the same counter-ion (NO-3). But, in leachates or sediments, there is rather a mixture of heavy metals. Thus, sorption studies have been carried out in a solution containing a mixture of heavy metals (copper, lead and chromium), in order to determine the behavior of the samples when there is a competition between several metals.
The competition between heavy metals to enter the reactive sites at the surface of the functionalized textile plays an important role in the adsorption efficiency of the material. Indeed, the competition between copper, chromium and lead is due to both the limited sites number and the size of the heavy metals. Sorption kinetics, as well as pH and ionic strength, could have an impact on the amount of heavy metals trapped. Fig. 7 shows that in a solution containing a mixture of heavy metals, the amount of ions trapped at the surface of the sample for each heavy metal is inferior to the amount trapped with the solution containing only the considered heavy metal. A maximum of copper can be trapped at pH 4.5 (33 mg Cu/g PP in the solution without sodium), and a maximum of chromium is trapped at pH 6.0 (13 mg Cr/g PP in the solution without sodium). At higher pH, the chromium concentration on PPg-AA-cysteine is too low to be detected by the atomic absorption spectrometer. At high pH (higher than 6), copper and chromium precipitate and thus the concentration of these heavy metals in the solution decreases. Then, less copper and chromium can be trapped by the functionalized textile and thus there are more and more free sites at the surface of the PP-g-AA-cysteine to trap lead. A maximum sorption capacity is observed at pH 8.0, with 63 mg Pb/g PP without sodium, which is much less important than the results obtained for the sorption of lead in a single metal containing solution. The presence of sodium also decreases the sorption efficiency of the PP-g-AA-cysteine. As observed for the single metal containing solutions, the amounts of copper and lead trapped decrease, due to the competition with sodium at the surface of PP-g-AA-cysteine. However, amounts of chromium trapped at pH 4.5 and 6.0 being similar whatever the sodium concentration, chromium adsorption seems not to be affected by the presence of sodium, but much more by the presence of copper and lead. This result could be explained by the small amount of chromium trapped. In the case of copper and lead, presence of sodium has an important effect on the sorption capacity: at pH 4.5, only 13 mg Cu/g PP are trapped instead of 33 mg Cu/g PP without sodium. 4. Conclusion
Fig. 6. Amount of lead, chromium and copper trapped by the PP-g-AA-cysteine as a function of the NaCl concentration in the contaminated solutions.
PP-g-AA-cysteine is a promising material in heavy metals depollution. Indeed, Cu(II), Pb(II) and Cr(III) can be trapped in interesting amounts from aqueous media using this material. Copper is the most interesting heavy metal because of the high
M. Vandenbossche et al. / Journal of Environmental Management 143 (2014) 99e105
amount that can be trapped and because of the adsorption kinetics: copper is trapped much quicker than the other heavy metals. EPR studies give information on the reactive functions involved in the metal coordination, and more precisely in the case of copper. Using this technique, it was shown that carboxylic acid groups, which are the most abundant chemical functions at the surface of the fibers, are involved in the trapping of copper. Thus, as these chemical groups can easily interact with sodium, it can explain the important decrease of the sorption capacity in the presence of NaCl: 45% for lead, 50% for copper and 71% for chromium. It was also observed that the adsorption capacity of the textile increases at high pH values: the material has an anionic structure at high pH, and thus the attraction between the anionic surface and the cationic heavy metals increases. But, generally, heavy metals precipitate at high pH (from pH 5e6), which does not allow the sorption of these heavy metals. This result is interesting as pH of marine sediments is about 8, thus this new material can be efficient to trap heavy metals at this pH. Acknowledgments The authors gratefully acknowledge the support of FEDER (Fonds Européen de Développement Régional), Nord-Pas-de-Calais region, and FUI (Fonds Unique Interministériel) for funding this work .We also would like to deeply acknowledge the support of the Up-Tex Competitiveness Cluster and all DEPOLTEX partners for helpful collaboration and discussion. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.jenvman.2014.05.002. References Brunauer, S., Deming, L.S., Deming, W.E., Teller, E., 1940. On a theory of the van der Waals adsorption of gases. J. Am. Chem. Soc. 62, 1723e1732. Chatterjee, A., Schiewer, S., 2014. Multi-resistance kinetic models for biosorption of Cd by raw and immobilized citrus peels in batch and packed-bed columns. Chem. Eng. J. 244, 105e116.
105
Chiou, M.S., Li, H.Y., 2003. Adsorption behavior of reactive dye in aqueous solution on chemical cross-linked chitosan beads. Chemosphere 50, 1095e1105. Chu, K.H., 2002. Removal of copper from aqueous solution by chitosan in prawn shell: adsorption equilibrium and kinetics. J. Hazard. Mater. 90, 77e95. Cobbett, C.S., 2000. Phytochelatins and their roles in heavy metal detoxification. Plant Physiol. 123, 825e832. Cooney, D.O., 1998. Adsorption Design for Wastewater Treatment. CRC Press. Dos Santos, V.C.G., Salvado, A.d.P.A., Dragunski, D.C., Peraro, D.N.C., Tarley, C.R.T., Caetano, J., 2012. Highly improved chromium (III) uptake capacity in modified sugarcane bagasse using different chemical treatments. Quim. Nova 35, 1606e 1611. Eggleton, J., Thomas, K.V., 2004. A review of factors affecting the release and bioavailability of contaminants during sediment disturbance events. Environ. Int. 30, 973e980. Foo, K.Y., Hameed, B.H., 2010. Insights into the modeling of adsorption isotherm systems. Chem. Eng. J. 156, 2e10. Gerente, C., Lee, V.K.C., Le Cloirec, P., McKay, G., 2007. Application of chitosan for the removal of metals from wastewaters by adsorption-mechanisms and models review. Crit. Rev. Env. Sci. Tec. 37, 41e127. Gouzy, A., Ducos, G., 2008. La connaissance des éléments-traces métalliques: un défi pour la gestion de l’environnement. Air Pur 75, 6e10. Klaassen, C.D., Liu, J., Choudhuri, S., 1999. METALLOTHIONEIN: an intracellular protein to protect against cadmium toxicity. Annu. Rev. Pharmacol. Toxicol. 39, 267e294. Langmuir, I., 1918. The adsorption of gases on plane surfaces of glass, mica and platinum. J. Am. Chem. Soc. 40, 1361e1403. Liu, W., Wang, T., Borthwick, A.G.L., Wang, Y., Yin, X., Li, X., Ni, J., 2013. Adsorption of Pb2þ, Cd2þ, Cu2þ and Cr3þ onto titanate nanotubes: competition and effect of inorganic ions. Sci. Total Environ. 456e457, 171e180. Maret, W., Larsen, K.S., Vallee, B.L., 1997. Coordination dynamics of biological zinc “clusters” in metallothioneins and in the DNA-binding domain of the transcription factor Gal4. Proc. Natl. Acad. Sci. 94, 2233e2237. Peisach, J., Blumberg, W.E., 1974. Structural implications derived from the analysis of electron paramagnetic resonance spectra of natural and artificial copper proteins. Arch. Biochem. Biophys. 165, 691e708. 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 Univ. Sci. A 10, 716e724. Singh, S.P., Tack, F.M.G., Verloo, M.G., 1998. Land disposal of heavy metal contaminated dredged sediments: a review of environmental aspects. Land Contam. Recl. 6, 149e158. Stronkhorst, J., Schot, M.E., Dubbeldam, M.C., Ho, K.T., 2003. A toxicity identification evaluation of silty marine harbor sediments to characterize persistent and nonpersistent constituents. Mar. Pollut. Bull. 46, 56e64. Vandenbossche, M., Casetta, M., Jimenez, M., Bellayer, S., Traisnel, M., 2014. Cysteine-grafted nonwoven geotextile: a new and efficient material for heavy metals sorption e part A. J. Environ. Manage 132, 107e112. Vandenbossche, M., Jimenez, M., Casetta, M., Bellayer, S., Beaurain, A., Bourbigot, S., Traisnel, M., 2013. Chitosan-grafted nonwoven geotextile for heavy metals sorption in sediments. React. Funct. Polym. 73, 53e59. Zenk, M.H., 1996. Heavy metal detoxification in higher plants e a review. Gene 179, 21e30.
