Effect of glutamic acid on copper sorption onto kaolinite - Batch experiments and surface complexation modeling

Chemosphere 178 (2017) 277e281

Contents lists available at ScienceDirect

Chemosphere journal homepage: www.elsevier.com/locate/chemosphere

Effect of glutamic acid on copper sorption onto kaolinite - Batch experiments and surface complexation modeling Lotfollah Karimzadeh*, Robert Barthen, Madlen Stockmann, Marion Gruendig, Karsten Franke, Johanna Lippmann-Pipke Helmholtz-Zentrum Dresden-Rossendorf, Institute of Resource Ecology, Permoserstr. 15, 04318 Leipzig, Germany

h i g h l i g h t s

g r a p h i c a l a b s t r a c t


Glutamic acid enhances copper mobility at alkaline pH.
Glutamic acid slightly enhances the Cu(II) sorption at low pH.
Glutamic acid sorption on kaolinite is weak and independent of pH.
CD-MUSIC model using monodentate complexation reactions described experiments well.

a r t i c l e i n f o

a b s t r a c t

Article history: Received 13 January 2017 Received in revised form 15 March 2017 Accepted 16 March 2017

High carbonate content of the European Kupferschiefer ore deposits is a challenge for acid copper leaching (pH  2). Therefore investigating the mobility behavior of Cu(II) under conditions related to an alternative, neutrophil biohydrometallurgical Cu(II) leaching approach is of interest. As glutamic acid (Glu) might be present as a component in the growth media, we studied its effects on the adsorption of Cu(II) onto kaolinite. The binary and ternary batch sorption measurements of Cu(II) and Glu onto kaolinite were performed in the presence of 10 mM NaClO4 as background electrolyte and at a pH range from 4 to 9. Sorption experiments were modeled by the charge-distribution multi-site ion complexation (CD-MUSIC) model by using single sorption site (≡SOH) and monodentate surface complexation reactions. Glu sorption on kaolinite is weak (<10%) and independent of pH. Furthermore, Glu slightly enhances the Cu(II) sorption at low pH but strongly hinders (up to 50%) the sorption at higher pH and therewith enhances copper mobility. The results of isotherms show that Cu(II)-Glu sorption onto kaolinite mimics the Freundlich model. The proposed CD-MUSIC model provides a close fit to the experimental data and predicts the sorption of Cu(II), Cu(II)-Glu and Glu onto kaolinite as well as the effect of Glu on Cu(II) mobility. © 2017 Elsevier Ltd. All rights reserved.

Handling Editor: Patryk Oleszczuk Keywords: Copper mobility Glutamic acid Kaolinite Sorption Surface complexation modeling

1. Introduction * Corresponding author. E-mail addresses: [email protected] (L. Karimzadeh), [email protected] (R. Barthen), [email protected] (M. Stockmann), [email protected] (M. Gruendig), [email protected] (K. Franke), [email protected] (J. Lippmann-Pipke). http://dx.doi.org/10.1016/j.chemosphere.2017.03.073 0045-6535/© 2017 Elsevier Ltd. All rights reserved.

Mobility and transport of copper in natural porous environments is strongly controlled by sorption and desorption processes. Low molecular weight organic ligands such as amino acids and carboxylic acids may enhance the sorption of copper onto solid/

278

L. Karimzadeh et al. / Chemosphere 178 (2017) 277e281

mineral surfaces via bridging between the surfaces and the metal ions (Fitts et al., 1999). On the other hand they can suppress the uptake of copper by forming mobile and strong complexes in solutions and/or competing with Cu2þ for the sorption surface sides (Fitts et al., 1999; Ikhsan et al., 2004; Wang et al., 2009). Previous studies (Fitts et al., 1999; Ikhsan et al., 2004; Yeasmin et al., 2014) have shown that amino acids can significantly alter the sorption of copper to (hydr)oxide surfaces. However, it is notable that due to the structural heterogeneity of amino acids, various sorption mechanisms have been introduced to explain their interaction with metals and mineral surfaces. According to Lambert (2008) amino acid sorption onto oxide surfaces occurs either via electrostatic interactions or by the formation of covalent or hydrogen bonds (outer sphere surface bonds). Ikhsan et al. (2004) studied the effect of aspartic acid on sorption of selected transition metals to kaolinite. They reported that aspartic acid slightly increases the sorption of copper at acidic pH due to electrostatic attraction to the permanent mineral surface charge sites. Whereas under neutral and alkaline conditions aspartic acid significantly hinders the Cu(II) sorption due to interactions with the variable surface charge sites. The type of amino acid and the mineral surface characteristics determine the nature of the sorption process, which is often a mix of different superimposed mechanisms (Lambert, 2008). When comparing the impact of different amino acids on metal sorption to a specific surface, both carboxyl groups and amino groups have a great influence on the sorption behavior (Benetoli et al., 2007; Wang and Lee, 1993; Yeasmin et al., 2014). Although several studies have discussed the effect of amino acids on metal sorption to mineral surfaces, influence of glutamic acid on copper sorption to clay mineral is still missing. Several surface complexation models have been introduced to explain Cu(II) sorption onto kaolinite both individually and in the presence of organic ligands (Gu and Evans, 2007; Hizal and Apak, 2005; Peacock and Sherman, 2005). Most of these are based on the Diffuse Layer Model (DLM) which was firstly proposed by Stumm and co-workers (Huang and Stumm, 1973; Stumm et al., 1970) and further improved by Dzombak and Morel (1990). Ikhsan et al. (2004) used a 2-site constant capacitance model with the formation of bidentate Cu(II) on the variable charge edge sites to explain Cu(II) sorption onto kaolinite. However, X-ray Absorption Fine Structure (XAFS) spectroscopy study indicated the formation of inner-sphere monodentate Zn-complexes on kaolinite surface edge sites (Nachtegaal and Sparks, 2004). Subsequent works by Hizal and Apak (2005) and Gu and Evans (2007) used monodentate complexes to predict Cu(II) sorption onto kaolinite. Small (2014) proposed the charge-distribution multi-site ion complexation (CD-MUSIC) model, introduced by Hiemstra and Van Riemsdijk (1996), for Cd sorption onto kaolinite to derive equilibrium constants for surface complexation reactions. She assumed a bidentate complexation reaction at low surface coverage, and both a bidentate and a monodentate complexation reaction at high surface coverage. The CD-MUSIC was initially developed for sorption on goethite, but has been successfully applied to many metal-oxide surfaces. The CD-MUSIC can accept multiple surface sites. In addition, the charge and the potential can be distributed over two additional planes in the double layer that extends from the surface into the solution. In comparisons to other models, the CD-MUSIC model has more options to fit experimental data. The CD-MUSIC model has become one of the most popular models to describe ion sorption on solid surfaces and has been applied to explain the sorption of organic acids onto goethite in recent years (Iglesias et al., 2010; Kersten et al., 2014; Small, 2014). As Glu might be present as a component in the growth media in neutrophil biohydrometallurgical Cu(II) leaching approach, we investigated its effects on the adsorption of Cu(II) onto kaolinite

and its implications for heavy metal mobility in soils in this study. Glu, HO2CCH2CH2CH(NH2)CO2H, as an acidic amino acid, possesses two carboxylic groups (a and g-carboxylate), as well as an amino group in its structure. We performed this work by conducting a series of batch sorption experiments in the pH range between 4 and 9. A predictive model was introduced to explain the sorption of Glu and its influence on copper sorption by kaolinite. The CD-MUSIC model was used to describe Cu2þ sorption onto kaolinite because of its flexibility to estimate parameters from data fitting over a wide range of conditions. 2. Materials and methods 2.1. Materials Kaolinite was obtained from Sigma-Aldrich Chemie GmbH, Germany. It was washed three times with 1 M NaCl (Merck, Germany) and three times with deionized water to avoid possible interferences caused by impurities. The copper stock solution (0.1 mM) was prepared by dissolving Cu(NO3)2$3 H2O (Merck, Germany) in ultrapure water (18 MU/cm). The exact Cu concentration was determined by Inductive Coupled Plasma Mass Spectroscopy (ICP-MS, Thermo Scientific ELEMENT XR) using a Cu standard solution (Merck, Germany) for calibration. Glu stock solution was provided by dissolving of L-Glutamic acid salt (ROTH, Germany) in ultrapure water. NaClO4 was used as a background electrolyte at a concentration of 0.01 M. The pH of the solutions was adjusted using 0.01 M HNO3 and 0.01 M NaOH. 2.2. Experimental Sorption experiments were performed in continuously stirred glass beakers containing 500 mL distilled water and 10 mM NaClO4 as the background electrolyte. Solutions were titrated first to an acidic initial pH (~4) using trace metal grade HNO3. Next, 500 mg kaolinite was added to the well-stirred solution. The suspension was pre-equilibrated for 24 h. After pre-equilibration, Glu and Cu(II) stock solutions were added to solutions to have binary sorption systems of 0.1 mM Glu, 0.1 mM Cu(II), and ternary system of 1:1 Cu(II):Glu. The pH of the suspensions with and without Glu were titrated upwards by increments of 0.5, varied from pH 4 to 9 by addition of 0.1 mM NaOH and left to equilibrate for 30 min. After stabilization of the pH within 0.02 log units, a 10 mL aliquot of the suspension was pipetted into an acid-washed 12 mL plastic vial sealed under N2 (g) to avoid interferences caused by CO2(g). The plastic tubes were agitated with a shaker (ROTATHERM) for 24 h under the laboratory conditions. Copper sorption isotherms were measured at a constant pH of 7, in the presence (0.1 mM) and absence of Glu. The concentration of copper ranged from 0 to 0.1 mM and concentration of kaolinite was 1 g/L. The 10 mL batch suspensions were prepared at 1, 5, 10, 50, and 100 mM of [Cu]total in 12 mL plastic vials. After stabilization of suspension pH, samples were sealed under N2 (g) and shaken continuously for 24 h. Afterwards the pH of the suspension was measured again, samples were centrifuged at 9000 rpm for 15 min, and filtered through 0.2 mm cellulous acetate syringe filters (Sartorius Stedim Biotech GmbH, Germany). All samples were acidified using trace metal grade HNO3 and analyzed for Cu using ICP-MS. Glu concentrations were determined by high-pressure liquidchromatography equipped with a fluorescence detector (Agilent, 1200 series). Samples were centrifuged (9000 rpm, 10 min) and filtered (0.2 mm) prior to analysis. Analysis was performed after online derivatization with o-phthalaldehyde (Sigma-Aldrich GmbH, Germany) and 2-Sulfanylethan-1-ol (Sigma-Aldrich GmbH,

L. Karimzadeh et al. / Chemosphere 178 (2017) 277e281

279

Germany) under alkaline conditions in 0.5 M potassium borate buffer. Separation was achieved on a reversed phase column (Zorbax Eclipse XDB-C18, Agilent, 4.6  150 mm, 5 mm) and corresponding guard column (Zorbax Eclipse XDB-C18, Agilent) in gradient elution mode. Eluent A: 40 mM phosphate buffer. Eluent B: Mixture of 45% acetonitrile, 45% methanol and 10% water. The flow rate was 0.6 mL/min, and the temperature was kept at 25  C. The excitation wavelength was set to 370 nm, and fluorescence was measured at 450 nm. All experiments were duplicated and the values averaged.

2.3. Sorption isotherms Equilibrium distribution of metal ions between the solid phase and liquid phase can be expressed by isotherm models such as Langmuir and Freundlich. The linear least-squares method to the linear isotherm equations has been applied to confirm the experimental data and isotherms (Shahmohammadi-Kalalagh and Babazadeh, 2014). Therefore in linear plots the amount of adsorbate (copper in mg) adsorbed per amount of adsorbent (kaolinite in g), qe (mg/g), is plotted against the equilibrium concentrations of adsorbate (Ce) in mg/L. The respective linear Freundlich isotherm is denoted by Equation (1),

  lnðqe Þ ¼ ln kf  ð1=nÞ lnðCe Þ

(1)

where, kf and n are Freundlich parameters which correspond to sorption capacity and intensity. The Freundlich parameters kf and n can be obtained from the slope and intercept of the plot between ln(qe) versus ln(Ce), respectively. The n value expresses the degree of nonlinearity of the relation between the amount of sorption and solution concentration at equilibrium. A value of n close to unity represents a linear isotherm, which indicates that the number of active sites remains constant up to saturation of sorption sites. In addition, sorption of uncharged aqueous species results in a linear isotherm, whereas species with electrostatic interactions show a non-linear sorption isotherm.

2.4. Speciation modeling and surface complexation modeling The aqueous chemical speciation of Cu(II), Glu and 1:1 Cu(II):Glu was calculated by the geochemical speciation code PHREEQC (version 3.3.8e11728) (Parkhurst and Appelo, 2013) using the thermodynamic database phreeqc.dat (Fig. 1). The additional stability constants for glutamate and Cu(II) -glutamate used in the model and their equilibrium reactions are presented in Table 1. For considering surface complexation reactions, the CD-MUSIC model was used to describe the sorption of Cu(II) and/or Glu on the surface of the kaolinite mineral. One type of surface hydroxyl groups (≡SOH) was chosen as sorption edge sites with formation of monodentate Cu(II) complexes on variable charge surface sites. For the estimation of the surface complexation parameter (SCP, namely log k) the experimental data from batch sorption experiments were fitted using the parameter estimation code UCODE_2014 (Poeter et al., 2014) coupled with PHREEQC. A weighted residuals model was used to fit the data. All relevant surface parameters and the surface complexation reaction for the considered binary and ternary surface complexes and the obtained log k values are given in Table 2.

Fig. 1. Speciation of Glu (a), Cu(II) (b), and 1:1 Cu(II):Glu (c) as a function of pH. [Cu]total ¼ [Glu]total ¼ 104 M in 0.01 M NaClO4, calculated with the geochemical code PHREEQC using the LLNL database and additional data from Table 1.

Table 1 Equilibrium constants (log k) and surface parameters used in the surface complexation modeling. ^SOH þ Hþ 4 ^SOHþ 2 ^SOH 4 ^SO þ Hþ 2 þ Glu þ H 4 HGl Glu2 þ 2Hþ 4 H2Glu Glu2 þ 3Hþ 4 H3Gluþ Cuþ2 þ Hþ þ Glu2 4 CuHGluþ Cuþ2 þ Glu2 4 CuGlu Cuþ2 þ 2Glu2 4 CuGlu2 2 Cuþ2 þ Glu2 þ OH 4 CuGluOH 2 Surface area (m /g) Density of sorption sites (sites/nm2) Capacitances (F/m2) a b c

3.96a 7.22a 9.51b 13.69b 15.96b 13.03b 8.52b 15.01b 1.85b 17.6c 5.2c 1.1, 5.0c

(Ikhsan et al., 1999). (Bregier-Jarzebowska, 2015). (Small, 2014).

3. Results and discussion 3.1. Sorption of glutamic acid The results of the sorption of Glu onto kaolinite in the pH range from 4 to 9 are presented in Fig. 2a. Glu uptake onto kaolinite is small (~5%) and independent of solution pH. This is in good

280

L. Karimzadeh et al. / Chemosphere 178 (2017) 277e281

3.2. Sorption of copper

Table 2 Predicted surface complexation reactions and isotherm parameters. Surface complexation reaction ^SOH ^SOH ^SOH ^SOH ^SOH ^SOH ^SOH ^SOH ^SOH

þ þ þ þ þ þ þ þ þ

log10 k

Cu2þ þ 2 H2O 4 ^SOHCu(OH)2 þ 2Hþ Cu2þ þ H2O 4 ^SOCuOH þ 2Hþ Cu2þ 4 ^SOCuþ þ Hþ Glu2 4 ^SOHGlu2 HGlu 4 ^SOH2Glu H2Glu 4 ^SOH3Glu CuGlu 4 ^SOHCuGlu 2 CuGlu2 2 4 ^SOHCuGlu2 CuHGluþ 4 ^SOCuHGlu þ Hþ

1.92 3.46 1.67 5.21 2.27 5.54 0.8 2.5 5.01

Freundlich Parameters

Cu

Cu-Glu

Kf n R

2.35 1.02 0.99

1.05 3.5 0.97

agreement with the results obtained by Yeasmin et al. (2014). They studied the sorption and desorption of selected amino acids including Glu onto kaolinite by means of ATR-FTIR spectroscopy and found that the amino group plays an important role in Glu sorption onto clay mineral surfaces. The formation of a protonated amino group may result in electrostatic uptake of glutamate to kaolinite. However, since the point of zero charge (pHpzc) of kaolinite is below pH 4.5 (Benetoli et al., 2007) it can be assumed that the kaolinite surface charge is mostly negative in the investigated pH range. Thus, the low sorption can be explained by electrostatic repulsion between kaolinite and the carboxylic groups of glutamate. Dashman and Stotzky (1982) reported that sorption of aspartic acid to kaolinite is also influenced by the type of the positively charged ions which saturate the clay minerals. The presence of copper resulted in a slightly increased sorption of Glu to kaolinite (Fig. 2a). This could be explained by the formation of a ternary complex where the positively charged cupric ion serves as a bridge between the negatively charged kaolinite surface and the Glu molecule.

Fig. 2. Results of measured (points) and modeled (lines) glutamic acid adsorption (a) and Cu(II) adsorption (b) on kaolinite.

For Cu(II) sorption onto kaolinite in the presence of 0.01 M NaClO4 as a function of pH we found a sigmoid sorption edge (Fig. 2b). The shape of the sorption edge is in good agreement with results of previous studies of Cu2þ sorption on kaolinite (Lund et al., 2008; Ikhsan et al., 1999; Jiang et al., 2010; Peacock and Sherman, 2005). In the absence of kaolinite and at the used Cu(II) concentration (0.1 mM) tenorite is predicted to be supersaturated at pH > 7. However, Lund et al. (2008) reported that at the same Cu concentration and in the presence of kaolinite sorption is the dominant mechanism of Cu2þ uptake and tenorite precipitation is negligible. In Fig. 2b the sorption behavior of Cu on kaolinite in the single sorbat system (absence of Glu) is compared to that of the binary system (presence of Glu). Sorption edge data show that Glu significantly affects Cu(II) sorption. When Glu is present above pH 5 the Cu(II) uptake is inhibited relative to the single-sorbat system. Speciation modeling (Fig. 1) shows that at pH > 5 a major fraction of aqueous copper occurs in complexed form with glutamate. As discussed before, the sorption of Glu on kaolinite is low. Thus, the major influence of Glu on copper sorption results from the formation of Cu-ligand complexes in solution. This observation is compatible with the conclusion from a previous study of the effect of aspartic acid on Cu(II) sorption onto kaolinite (Ikhsan et al., 2004). By contrast, under slightly acidic conditions (pH 4), Glu slightly enhances Cu(II) uptake onto kaolinite. This can be interpreted as sorption of Cu-glutamate complexes on the sorption surface edge sites of kaolinite. Fitts et al. (1999) used spectroscopic methods to study the structure and bonding of Cu(II)-glutamate complexes at the g-Al2O3-water interface. They found that in acidic conditions, Glu forms a bridge between metal ions and the alumina surface and consequently increases metal uptake.

Fig. 3. Effect of initial Cu(II) solution concentrations (Ci) on kaolinite adsorption(qe); lines are modeled results and points are experimental data (a), and Freundlich isotherms for Cu, and Cu-Glu adsorption (b).

L. Karimzadeh et al. / Chemosphere 178 (2017) 277e281

3.3. Sorption isotherms In Fig. 3a the observed linear relationship between initial copper concentration in solution (mg/L) and the mass of copper adsorbed on kaolinite (mg/g) is displayed. The equilibrium distribution of the copper ions between the solid phase and liquid phase can be expressed by isotherm models such as Langmuir and Freundlich (Equation (1)). In Fig. 3b the linear plots of amount of adsorbate (copper) adsorbed per amount of adsorbent (kaolinite) against the equilibrium concentrations of adsorbate (Ce) are presented. The results show that Cu(II) sorption onto kaolinite obeys the Freundlich model with the correlation coefficient of 0.99 for Cu(II) and 0.97 for Cu(II)-Glu. The Freundlich parameters kf and n are obtained from the slope and intercept of the plot between ln(qe) versus ln(Ce), and results are presented in Table 2. 3.4. Surface complexation modeling Experimental data revealed that whether or not copper was present in the system, Glu adsorbed only slightly to kaolinite. The low affinity of glutamate, as well as the weak influence of cupric ions on glutamate sorption to kaolinite surfaces, is well fitted and presented in the results of modeling (Fig. 2a) with the reaction constant k for the proposed surface complexation reactions being the only fitted parameter (Table 2). Furthermore, modeling of the Cu(II) sorption edge data using the surface complexation reactions, provides a close fit to the experimental data (Fig. 2b) and predicted the effect of Glu on metal sorption. At high pH (pH > 5), when Glu was present, the model predicted a decrease in total Cu(II) sorption due to the formation of Cu(II)-Glu complexes in solution. In addition the slightly enhanced Cu(II) sorption at low pH (pH of 4) was also displayed in the results of the fit. Surface complexation modeling also closely predicted the sorption isotherm obtained at pH of 7 (Fig. 3a), demonstrating that the model well describes the sorption behavior of kaolinite over a wide range of Cu(II):Glu concentrations. 4. Conclusion Copper sorption in the absence of Glu was dependent on pH and initial copper concentration. Sorption of Glu was weak and independent of pH. Furthermore, Glu reduces the copper sorption onto kaolinite at alkaline pH and therewith enhances copper mobility in the presence of kaolinite due to the formation of Cu(II)-ligand complexes in solution. However, under acidic pH Glu forms a bridge between Cu2þ and kaolinite sorption surface sides, and consequently decreases the mobility of copper. In addition, presence of Cu(II) in the sorption system slightly elevated the sorption of Glu by kaolinite due to the formation of ternary surface complexes in which Cu2þ bridges between organic ligand and the surface site. The application of the CD-MUSIC model successfully describes the experimental data. Both the sorption of copper and the copper-Glu complex could be successfully described using Freundlich isotherms. These results show that copper mobility might be substantially increased in the growth media containing glutamic acid, especially at alkaline conditions. The obtained thermodynamic data for the ternary system copper, Glu and kaolinite are substantial for the quantitative geochemical modeling of neutrophil bioleaching of copper from e.g. Kupferschiefer ore deposits in northeastern Europe. Acknowledgments We gratefully acknowledge financial support by the German Federal Ministry of Education and Research (BMBF), project ref. no. 033RF001, and by the Agence Nationale de la Recherche (ANR)

281

(ANR-13-RMNP-0006), France. We also would like to thank Claudia ۧler for her support related to the ICP-MS analyses. Scho Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.chemosphere.2017.03.073. References Benetoli, L.O., de Souza, C.M., da Silva, K.L., de Souza Jr., I.G., de Santana, H., Paesano Jr., A., da Costa, A.C., Zaia, C.T., Zaia, D.A., 2007. Amino acid interaction with and adsorption on clays: FT-IR and Mossbauer spectroscopy and X-ray diffractometry investigations. Orig. Life Evol. Biosph. 37, 479e493. Bregier-Jarzebowska, R., 2015. Mixed-ligand complexes of copper(II) ions with Lglutamic acid in the systems with triamines and non-covalent interaction between bioligands in aqueous solution. Open Chem. 13, 113e124. Dashman, T., Stotzky, G., 1982. Adsorption and binding of amino-acids on homoionic montmorillonite and kaolinite. Soil Biol. Biochem. 14, 447e456. Dzombak, D.A., Morel, F.M.M., 1990. Surface Complexation Modeling: Hydrous Ferric Oxide, vol. 1990. John Wiley & Sons, New York. Fitts, J.P., Persson, P., Brown, G.E., Parks, G.A., 1999. Structure and bonding of Cu(II)glutamate complexes at the g-Al2O3-water interface. J. Colloid Interface Sci. 220, 133e147. Gu, X., Evans, L.J., 2007. Surface complexation modelling of Cd(II), Cu(II), Ni(II), Pb(II) and Zn(II) adsorption onto kaolinite. Geochem. Cosmochem. Acta 72, 267e276. Hiemstra, T., Van Riemsdijk, W.H., 1996. A surface structural approach to ion adsorption: the charge distribution (CD) model. J. Colloid Interf. Sci. 179, 488e508. Hizal, J., Apak, R., 2005. Modeling of copper(II) and lead(II) adsorption on kaolinitebased clay minerals individually and in the presence of humic acid. Colloid Interface Sci. 295, 1e13. Huang, C.-P., Stumm, W., 1973. Specific adsorption of cations on hydrous g-Al2O3. Colloid Interface Sci. 43, 409e420. pez, R., Gondar, D., Antelo, J., Fiol, S., Arce, F., 2010. Adsorption of Iglesias, A., Lo MCPA on goethite and humic acid-coated goethite. Chemosphere 78, 1403e1408. Ikhsan, J., Johnson, B.B., Wells, J.D., 1999. A comparative study of the adsorption of transition metals on kaolinite. J. Colloid Interface Sci. 217, 403e410. Ikhsan, J., Wells, J.D., Johnson, B.B., Angove, M.J., 2004. The effect of aspartic acid on the binding of transition metals to kaolinite. J. Colloid Interface Sci. 273, 6e13. Jiang, M.Q., Jin, X.Y., Lu, X.Q., Chen, Z.L., 2010. Adsorption of Pb(II), Cd(II), Ni(II) and Cu(II) onto natural kaolinite clay. Desalination 252, 33e39. Kersten, M., Tunega, D., Georgieva, I., Vlasova, N., Branscheid, R., 2014. Adsorption of the herbicide 4-Chloro-2-methylphenoxyacetic acid (MCPA) by goethite, 48, 11803e11810. Lambert, J.F., 2008. Adsorption and polymerization of amino acids on mineral surfaces: a review. Orig. Life Evol. Biosph. 38, 42. Lund, T.J., Koretsky, C.M., Landry, C.J., Schaller, M.S., Das, S., 2008. Surface complexation modeling of Cu(II) adsorption on mixtures of hydrous ferric oxide and kaolinite. Geochem Trans. 9, 9. Nachtegaal, M., Sparks, D.L., 2004. Effect of iron oxide coatings on zinc sorption mechanisms at the clay-mineral/water interface. Colloid Interface Sci. 276, 13e23. Parkhurst, D.L., Appelo, C.A.J., 2013. Description of input and examples for PHREEQC version 3da computer program for speciation, batch-reaction, one-dimensional transport, and inverse geochemical calculations. book 6. In: U.S. Geological Survey Techniques and Methods, Denver, Colorado, p. 497. Peacock, C.L., Sherman, D.M., 2005. Surface complexation model for multisite adsorption of copper(II) onto kaolinite. Geochim. Cosmochim. Acta 69, 3733e3745. Poeter, E.P., Hill, M.C., Lu, D., Tiedeman, C.R., Mehl, S., 2014. UCODE_2014, with New Capabilities to Define Parameters Unique to Predictions, Calculate Weights Using Simulated Values, Estimate Parameters with SVD, Evaluate Uncertainty with MCMC, and More. Integrated Groundwater Modeling Center, p. 189. Shahmohammadi-Kalalagh, S., Babazadeh, H., 2014. Isotherms for the sorption of zinc and copper onto kaolinite: comparison of various error functions. Int. J. Environ. Sci. Technol. 11, 111e118. Small, M.A., 2014. Surface Complexation and Reactivity of Kaolinite and Gibbsite with Cadmium: a Multi-method Approach. University of California, Mercwd, p. 100. Stumm, W., Huang, C.P., Jenkins, S.R., 1970. Specific chemical interaction affecting the stability of dispersed systems. Croat. Chem. Acta 42, 223e245. Wang, X.C., Lee, C., 1993. Adsorption and desorption of aliphatic-amines, aminoacids and acetate by clay-minerals and marine-sediments. Mar. Chem. 44, 1e23. Wang, Y., Chena, J., Cuia, Y.X., Wang, S.Q., Zhou, D.M., 2009. Effects of lowmolecular-weight organic acids on Cu(II) adsorption onto hydroxyapatite nanoparticles. J. Hazard. Mater. 162, 1135e1140. Yeasmin, S., Singh, B., Kookana, R.S., Farrell, M., Sparks, D.L., Johnston, C.T., 2014. Influence of mineral characteristics on the retention of low molecular weight organic compounds: a batch sorption-desorption and ATR-FTIR study. J. Colloid Interface Sci. 432, 246e257.