Competitive Removal of Heavy Metals from AqueousSolutions by Montmorillonitic and Calcareous Clays

Ali Sdiri & Teruo Higashi & Rochdi Chaabouni &Fakher Jamoussi

Received: 4 July 2011 /Accepted: 11 August 2011# Springer Science+Business Media B.V. 2011

Abstract A batch sorption method was used to studythe removal of few toxic metals onto the Late Cretaceousclays of Aleg formation (Coniacian–Lower Campaniansystem), Tunisia, in single, binary and multi-componentsystems. The collected clay samples were used asadsorbents for the removal of Pb(II), Cd(II), Cu(II) andZn(II) from aqueous solutions. Results show that thenatural clay samples were mainly composed of silica,alumina, iron and magnesium oxides. N2-adsorptionmeasurements indicated mesoporous materials withmodest specific surface area of <71 m2/g. Carbonateminerals were the most influencing parameters forheavy metal removal by natural clays in both singleand multi-element systems. The affinity sequence wasPb(II)>Cu(II)>Zn(II)>Cd(II) due to the variable physi-

cal properties of the studied metals. The maximumadsorption capacity reached 131.58 mg/g in singlesystems, but decreased to <50.10 mg/g in mixedsystems. In single, binary and muti-element systems,the studied clay samples removed substantial amountsof heavy metals, showing better effectiveness than therelevant previous studies. These results suggest that thestudied clay samples of the Late Cretaceous clays fromTunisia can be effectively used as natural adsorbents forthe removal of toxic heavy metals in aqueous systems.

Keywords Heavymetals .Wastewaters . Clays .

Competitive adsorption . Relative binding strength

1 Introduction

The present study has been carried out to evaluate theapplicability of natural clays in the removal of severalmetal ions (Pb(II), Cd(II), Cu(II) and Zn(II)) fromaqueous solutions. Batch adsorption technique was usedfor the removal experiments in single, binary and multi-component systems. The use of natural adsorbents isparticularly beneficial for the development of costeffective process for heavy metal removal from waste-waters. In this regards, multiple researchers studied theeffectiveness of various type of clay in removing heavymetals from aqueous solutions, including kaolinite (Sariet al. 2007a; Schaller et al. 2009), Celtek clay (Sari etal. 2007b), bentonite (Ulmanu et al. 2003; Kaya and

Water Air Soil PollutDOI 10.1007/s11270-011-0937-z

A. Sdiri (*) : T. HigashiGraduate School of Life and Environmental Sciences,University of Tsukuba,1-1-1 Tennodai,Tsukuba 305-8572, Japane-mail: alisdiri@yahoo.fr

R. ChaabouniLaboratory of Water, Energy and Environment (L3E),National School of Engineers, University of Sfax,Sfax 1173-3038, Tunisia

F. JamoussiNational Center for Material Sciences (CNRSM),Borj Cédria,Soliman 273-8020, Tunisia

Hakan Ören 2005), sepiolite (Guerra et al. 2010),montmorillonite (Lin and Juang 2002) and othernatural adsorbents (Al-Degs et al. 2006).

In Tunisia, the use of natural clays as local adsorbentsfor the removal of heavy metals has been extensivelyinvestigated (Chaari et al. 2008; Eloussaief et al. 2009,2011; Eloussaief and Benzina 2010; Ghorbel-Abid etal. 2010; Sdiri et al. 2011a). Among the studied naturalclayey deposits, the Coniacian–Lower Campanianclays of the Aleg formation, discovered in Jebel Oumel Aleg to the north of Gafsa area (south of Tunisia),has the required technical specifications for use invarious industrial applications, especially for heavymetal removal (Chaari et al. 2008; Sdiri et al. 2010a).However, most of the studies mentioned above weredirected to study the removal of heavy metals in singleelement systems. Actually, contaminated wastewaterscontain more than one pollutant such as Pb(II), Cd(II),Cu(II) and Zn(II) cations that have to be removedbefore the final discharge in the receiving ecosystems.Adsorption onto cost-effective natural materials is themost commonly utilized technique due to its simplicitywhen compared to other methods like ion exchangeand precipitation. In this study, we attempted to usenatural clay samples without any pre-treatment in orderto decrease the cost of the removal process. Moreover,the advantages of Tunisian clay in term of competitiveadsorption of various metal cations were demonstratedas a continuation of our effort on valorizing the naturalgeological resources of Tunisia. The effects of impuritiesas well as the competitive cations on the adsorption ofheavy metals were also emphasized. The importance ofunderstanding the mechanisms involved in heavy metalsremoval from wastewaters lied on deciphering themutual interactive behavior of both clay and metalremoved, and between metal cations in mixed systems.

2 Geological Setting

Natural clay samples used as adsorbent were collectedfrom the Aleg formation (Coniacian–Early Campanian)in the Upper Cretaceous outcroppings exposed in theAtlasic domain (south of Tunisia). The Aleg formationconsists of a thick series of clayey deposits and graymarl intercalated between the top of Zebbag and thelower part of Abiod limestones. In Jebel Oum el Aleg tothe northeastern part of Gafsa area (south of Tunisia),Burollet (1956) described Aleg formation as a thick

clay deposits with intercalated beds of marl andlimestones. In the Gabes region in southeasternTunisia, the Coniacian–Early Campanian systemextends over a large surface and presents a greatvariation in its thickness because of the active tectonicevents that affected the whole sector during thedeposition of the Aleg clays (Fig. 1). Mejri et al.(2006) metioned that the thick shale of “Upper Aleg”in eastern Tunisia resulted from the sedimentation afterstrong submarine erosion of the previous seriescombined with discrete tectonic movements anddifferential subsidence related to salt migration. In theJebel Aidoudi study site, the stratigraphic successionshows an argillaceous complex admitting carbonateintercalations in the lower part; a marly and calcareouscomplex and a clay sequence with fibrous gypsum.The total thickness of this formation exceeds 320 m(Abdeljaouad 1983).

3 Materials and Methods

3.1 Materials

Two natural clay samples (S and Y) were collected fromthe Coniacian–Early Campanian outcroppings of Gabes(Y sample) and Gafsa (S sample) areas (south ofTunisia) (Fig. 1). The preparation of the adsorbentfrom natural clays was limited to the removal ofcarbonate minerals and organic matter to concentratethe clay minerals, and to improve their sorptionproperties with an economic treatment cost. Carbonateswere removed using acetic acid (1 mol/l) in a waterbath at 80°C. Organic matter was oxidized by treatingthe carbonate-free samples with 30% H2O2. Bothoriginal and treated clay materials were dispersed inwater and saturated with 1 M NaCl solution for threetimes under continuous stirring. The obtained Na+

homoionic clay samples were then dialyzed severaltimes to remove chloride (Darder et al. 2005; Jarraya etal. 2010). Finally, Na+ homoionic clay materials weredried at 105°C for 24 h and then crushed to collect thedesired particle size ranges of less than 210 μm.Hereafter, we use the abbreviations RS and TS, RYandTY to refer to the original and treated S and Y clays,respectively.

All the chemicals used were of analytical reagentgrade supplied by Wako Pure Chemical Industries,Ltd. (Japan). Ultrapure water, produced with a Milli-Q

Water Air Soil Pollut

system (Millipore Corp., France), was used through-out the experimental studies. Stocks of copper,cadmium, lead and zinc solutions (1,000 mg/l) wereprepared by dissolving appropriate amounts ofCuCl2·2H2O, CdCl2·2.5H2O, PbCl2 and ZnCl2 in1,000 ml of ultrapure water. Working standards wereprepared by dilution of the stock solution. 0.1 M HCland 0.1 M NaOH solutions were used for pHadjustment.

3.2 Batch Adsorption

Batch adsorption is a simple technique commonlyutilized to assess the adsorptive capacities of naturaland synthetic sorbents. Aside from the fact that it is easyto manipulate, it also helps in obtaining importantinformation about the efficiency of a given sorbent to

remove the studied solute in static conditions. In thepresent study, 1 g/l of each original (RS and RY)and treated (TS and TY) adsorbent was placed in apolypropylene centrifuge tube containing metal ionsolution of known concentration and pH. The pH ofmetal solution was adjusted prior to mixing withthe adsorbent with 0.1 M HCl and 0.1 M NaOH.The tubes were shaken at 200 rpm for 60 min toreach equilibrium. The suspensionwas then centrifuged,and 10 ml supernatant were withdrawn and filteredwith a 0.45-μm syringe driven filter (Millex-LH, PTFE,Millipore Corp., Ireland). The filtered solutions werestored at 4°C till the measurement of the desired metal(Pb(II), Cd(II), Cu(II) and Zn(II)) concentrationusing an ICP-AES Instruments, Optima 7300 DVseries (PerkinElmer Inc., Japan). Adsorption of theselected metal ions was carried out with concentrations

Fig. 1 Distribution of theLate Cretaceous formationsand locations of thecollected samples (star)(modified after Ben Haj etal. 1985)

Water Air Soil Pollut

of Cd(II) 10mg/l, Zn(II) 20 mg/l, Cu(II) 40mg/l, and Pb(II) 60 mg/l under the operating pH of 6, and agitationspeed of 200 rpm within the equilibrium time of 60 minat 25°C for single and multi-component systems. Theamount of removed heavymetal was calculated from thedifference between initial and final concentrations. Allexperiments were run in triplicate.

4 Results and Discussions

4.1 Characterization of Clay Samples

Chemical compositions of the original and treatedsamples were determined by subjecting the pelletedclay powders to an electron microprobe equippedwith an X-ray dispersive spectrometer (JXA8621Superprobe; JEOL, Japan). The obtained resultsshowed that the original (RS and RY) and treated(TS and TY) clays were mainly composed of silica,aluminum and iron oxides (Table 1). The results alsoindicated that RS clay, collected from Gafsa area,contained a high amount of calcium carbonatesreaching 26.67%, while the RY sample contained

more SiO2, Al2O3, and Fe2O3. Based on the chemicalproperties, it was expected that the studied claysamples show variable adsorptive behaviors. Highcarbonate contents may enhance Pb(II), Cu(II) and Zn(II) removal. However, the high SiO2, Al2O3 andFe2O3 contents in RY and TY may suggest higherremoval amounts of metal cations due to the presencealuminol (Al–O) and silanol (Si–O) groups at thesurface of both RYand TY clay samples.

Mineralogical analysis was carried out with an X-ray diffractometer (RAD-X; Rigaku Intl. Corp.,Japan) using Cu kα radiation (40 kV, 25 mA). Theobtained diffractogram of RS showed the characteristicpeaks of calcite at 3.03Å (100), dolomite at 2.89Å (006)and the additional peak of quartz prevailing at 3.33–3.34Å (101) (Sdiri et al. 2010a). For the original Ysample (RY), the main peak of smectite occurring near12.2Å was subordinated by the characteristic peak ofquartz (Fig. 2).

N2 adsorption methodology was used for porositymeasurement using Coulter instrument (SA3100,Beckman Coulter, USA). The obtained results showedthe specific surface areas of the original and treatedclay samples (less than 210-μm-sized powders) that

Elements and properties Original samples Treated samples

RS RY TS TY

SiO2 40.096 57.567 57.144 61.759

Al2O3 16.320 21.138 22.495 21.359

Fe2O3 7.386 10.181 9.891 10.093

CaO 26.667 2.370 2.372 0.324

MgO 4.667 3.615 2.751 1.391

TiO2 1.158 1.111 1.472 1.249

K2O 3.060 1.744 3.604 1.635

Na2O nd 0.605 – 1.608

P2O5 0.429 0.901 0.015 0.019

Cr2O3 0.067 0.046 – –

MnO 0.083 nd 0.130 0.090

SrO 0.050 0.031 0.047 0.024

CuO 0.002 0.003 nd nd

ZnO 0.018 0.020 0.024 0.023

CdO nd nd nd 0.333

PbO nd nd 0.005 nd

SBET (m2/g) 17.843 71.940 10.222 59.141

Fine fraction of <2 μm (%) 42.751 88.533 – –

CEC (cmol(+)/kg) 37.450 36.871 20.171 29.663

Table 1 Physico-chemicalproperties of the studiedclay samples (% by weight)

nd not detected

Water Air Soil Pollut

were estimated to be 17.84, 71.94, 10.22 and 59.14 m2/gfor RS, RY, TS and TY, respectively (Table 1).

The adsorbent was also characterized using an FT-IR spectrophotometer (FT-720; Horiba Ltd., Japan) atroom temperature. The obtained FTIR spectra of thebulk (RS and RY) clay samples as well as theirrespective decarbonated samples (TS and TY) showedthe stretching vibrations of the surface hydroxylgroups (Si–Si–OH, or Al–Al–OH) near 3,694.94 and3,619.73 cm−1. The other vibration at 1,635.34 cm−1

was attributed to the bending of adsorbed water (Fig. 3).The stretching bands near 1,435.34, 871.67 and711.60 cm−1 were assigned to calcite (Sdiri et al.

2010b; 2011a). The removal of carbonate mineralsby acid treatment led to the elimination of the above-mentioned bands of calcite in TS and TY spectra.In the low frequency range (1,200 to 650 cm−1), themaximum absorption of silicate minerals was observedat 998.95 and 991.23 cm−1 for RS and RY, respectively.Other vibrations at 912.16, 796.46, 779.10 and692.32 cm−1 were also characteristic of silicateminerals (Sdiri et al. 2011a).

4.2 Adsorption Experiment

4.2.1 Adsorption in Single-Element System

The adsorption of the selectedmetal ions was carried outwith concentrations of Cd(II) 10 mg/l, Zn(II) 20 mg/l,Cu(II) 40 mg/l, and Pb(II) 60 mg/l under the operatingpH of 6 and agitation speed of 200 rpm within theequilibrium time of 60 min. As shown in Fig. 4, a highremoval efficiency of Pb(II) was achieved by RS dueto the precipitation as lead carbonate, but the removalcapacity decreased to 18.25 mg/g Pb(II) for TS becauseof the removal of carbonate minerals. Similar resultswere found by Sdiri et al. (2011b), when studying theeffects of impurities on the removal of heavy metals bynatural limestones. They stated that due to the presenceof carbonates minerals, Pb removal process wasgoverned by the precipitation of lead carbonate(PbCO3). The low precipitation pH of PbCO3 (pH5.3) and the solubility product constant (Ksp=7.4×10−14 at 25°C) would explain this phenomenon. AbovepH 5.3, solid-phase PbCO3 should form, leading to ahigh sorption capacity, since the equilibrium pHincreased to 7.7 after the addition of carbonaceousclay (RS sample). The same behaviour was observedwith Cu(II) and Zn(II) removal by S sample (RS andTS) with a decrease from 19.55 to 10.96 mg/g andfrom 15.22 to 9 mg/g for Cu(II) and Zn(II) removal,respectively, but not for Cd(II). The amount of Cd(II)removed was 5.58, 5.73, 9.03 and 9.42 mg/g by RS,TS, RYand TY, respectively (Fig. 4). This indicates thatboth original and carbonate free samples removedsimilar amounts of Cd(II).Therefore, it is possible thatthe removal of Cd(II) may not be influenced by thepresence of carbonate minerals. As for RYand TY, theyremoved similar amounts of Pb(II), Cd(II), Cu(II) andZn(II) because of the small amounts of carbonate in theoriginal sample, and therefore the similar physico-chemical properties of both forms (Table 1).Fig. 3 Infrared spectra of natural clay samples

Fig. 2 X-ray diffractograms of randomly oriented powder (Ssmectite, K kaolinite, Q quartz, C calcite, D dolomite; spacingis in angstrom)

Water Air Soil Pollut

4.2.2 Adsorption in Binary System

Figure 5 shows the removal quantities of Pb(II), Cd(II),Cu(II) and Zn(II) by the original and treated claysamples at pH 6 for all possible metal combinations inbinary systems. The results show that Pb(II) has thehighest removal efficiency. In Pb–Cd systems, morethan 24.87 mg/g of Pb(II) were removed, indicating thatthe presence of Cd(II) has a poor inhibitory effect on Pb(II) adsorption onto original and treated clays (Table 2).The removal amounts of Pb(II) in the presence of Cd(II) were 48.04, 34.61, 24.87 and 35.2 mg/g for RS,RY, TS and TY samples, respectively. The carbona-ceous clay (RS) removed 48.04 mg/g of Pb(II), butdramatically decreased for its carbonate-free form (TS)(Fig. 5). Thus, the presence of carbonates in RS samplehas an enhanced Pb(II) removal because of theprecipitation of PbCO3. In contrast, smectitic claysamples (RY and TY) showed comparable removalcapacities because of their similar physico-chemicalproperties (e.g., specific surface area and chemicalcomposition). Such behavior was observed in allexperiments regardless of the metal removed and thestudied mixture. The removal efficiency of Cd(II)decreased substantially in the presence of Pb(II)compared with single element system (Figs. 4 and 5).Therefore, Pb(II) strongly depressed Cd(II) adsorption

with an inhibitory effect on Cd(II) removal (defined as

IE (%) =CdðCdÞ�CdðPb�CdÞ

CdðCdÞ, where Cd(Cd), Cd(Pb–Cd) are Cd

(II) removal rate (mg/g) in the single and the Pb–Cdsystems, respectively): 79.39%, 60.01%, 72.66% and52.76% for RS, RY, TS and TY samples, respectively.Inversely, the inhibitory effect of Cd(II) on Pb(II)removal occurred to a much lower extent because ofthe higher selectivity for Pb(II) combined with the lowcompetitive ability of Cd(II) with regards to Pb(II). Itwas interesting to note that, except for RS sample, thehighest removal amounts of Pb(II) were observed inPb–Cd system.

The interactive behavior of Pb–Cu system showed thatthe studied clay samples removed a higher amount of Pb(II) than Cu(II), proving their higher affinity to the former.Similar results were reported by Usman (2008), whostudied the competitive adsorption of heavy metals ontoEgyptian smectitic soils. However, the inhibitory effectof Cu(II) on Pb(II) removal was higher in all cases,except for the TS sample. It was found that the presenceof Cu(II) strongly prevented the removal of Pb(II), withan inhibitory effect (IE) of 58.70%, 34.41% and 46.12%for RY, TS and TY samples, respectively (Table 2).

In the presence of Zn(II), the studied samplesremoved higher amounts of Pb(II) than in Pb–Cusystem. This meant that Zn(II) was less effective than

37.85

18.25

36.16

0

10

20

30

40

RS RY TS TY

Pb

rem

oval

(m

g/g)

Pb

19.5516.63

10.96

16.78

0

10

20

30

40

RS RY TS TY

Cu

rem

oval

(mg/

g)

Cu

5.589.03

5.739.42

0

10

20

30

40

RS RY TS TY

Cd

rem

oval

(m

g/g)

Cd

15.2213.37

9.00

14.40

0

10

20

30

40

RS RY TS TYZ

n re

mov

al(m

g/g)

Zn

Fig. 4 Single-element adsorption onto natural clay samples

Water Air Soil Pollut

Cu(II) in preventing Pb(II) adsorption to the bindingsites. Hence, it is plausible to account for lowcompetition between Zn(II) and Pb(II), as confirmedby their different physico-chemical properties (Table 3).

In the Cu–Cd system, the removal of Cd(II) wasstrongly affected by Cu(II). For instance, the presence ofCu(II) decreased Cd(II) removal by 82.62%, 68.77%,77.84% and 66.98% for RS, RY, TS and TY, respectively.On the contrary, Cd(II) was the metal with less influenceon the retention Cu(II) probably because of the existenceof the specific sites for Cu(II) adsorption, so that when itis present, these sites would not be occupied by Cd(II)(Echeverria et al. 1998).

In the Cu–Zn system, despite their similar ionicradii, the presence of Cu(II) affected the removal ofZn(II) to the same extent as that for Cd(II). This wasexpected since Cu(II) was predominantly adsorbed by

all the studied samples over Cd(II) and Zn(II) becauseof its higher relative binding strength and lowerionization potential (Table 3). As for the Cd–Znsystem, it was observed that Zn(II) highly affectedthe removal of Cd(II) while Cd(II) slightly preventedthe removal of Zn(II) cations, except for RS sample.Mahamadi and Nharingo (2010) found that, becauseof their similar physico-chemical properties, Cd(II)and Zn(II) should compete with the same order ofmagnitude. However, this was not the case of thepresent clay samples. Such behavior resulted from thecombination of various factors, including hydrolysisconstant and relative binding strength, among others.In summary, the studied clay samples had the highestaffinity for Pb(II) ions and the least affinity for Cd(II)ions. The adsorption affinity order was Pb(II)>Cu(II)>Zn(II)>Cd(II), showing a clear dependence to the

1.153.61 1.57

4.45

0

10

20

30

40

50

RS RY TS TY

Cd

rem

oval

(m

g/g)

Pb-Cd

48.04

34.61

24.87

35.20

0

10

20

30

40

50

RS RY TS TY

Pb

rem

oval

(m

g/g)

Pb-Cd

8.6211.14

6.6210.95

0

10

20

30

40

50

RS RY TS TYC

u re

mov

al (

mg/

g)

Pb-Cu15.6319.90

11.97

19.48

0

10

20

30

40

50

RS RY TS TY

Pb

rem

oval

(m

g/g)

Pb-Cu

3.666.64

3.057.03

0

10

20

30

40

50

RS RY TS TY

Zn

rem

oval

(m

g/g)

Pb-Zn

50.10

28.60

21.14

28.17

0

10

20

30

40

50

60

RS RY TS TY

Pb

rem

oval

(m

g/g)

Pb-Zn

Fig. 5 Adsorption of Pb(II), Cd(II), Cu(II) and Zn(II) by natural clays in binary systems

Water Air Soil Pollut

physico-chemical properties of the studied metal to ahigher extent than the adsorbent characteristics.Similar results were found by Pagnanelli et al.(2003) and Qin et al. (2006).

4.2.3 Adsorption in Ternary System

Competitive sorption experiments were also performedby simultaneous removal of three metals under the sameexperimental conditions described above. The resultsindicated that in ternary systems, the removed amountsof metal cations decreased when compared with single

and binary systems. This was expected because of thegreater suppression of metal led to the increase inelectrostatic repulsion. The results also showed higheraffinity of the studied clay samples to Pb(II) and Cu(II)with the sequence of Pb(II)>Cu(II)>Zn(II)>Cd(II). Asimilar sequence has been reported for the competitiveadsorption of several metals onto different soil samplesby Usman (2008), who found a sequence of Pb(II)>Cu(II)>Zn(II)>Cd(II), confirming that Pb(II) had thehighest affinity than other metals. Comparing the dataobtained for lead, cadmium, copper and zinc, a strongrelationship between metal removal efficiency and their

0.97 2.82 1.27 3.11

0

10

20

30

40

50

RS RY TS TY

Cd

rem

oval

(m

g/g)

Cd-Cu16.97 14.869.48

14.98

0

10

20

30

40

50

RS RY TS TY

Cu

rem

oval

(m

g/g)

Cd-Cu

1.874.61

2.155.24

0

10

20

30

40

50

RS RY TS TY

Cd

rem

oval

(m

g/g)

Cd-Zn

4.18

10.615.06

11.07

0

10

20

30

40

50

RS RY TS TY

Zn

rem

oval

(m

g/g)

Cd-Zn

1.755.00

2.505.25

0

10

20

30

40

50

RS RY TS TY

Zn

rem

oval

(m

g/g)

Cu-Zn17.5612.85

8.9112.48

0

10

20

30

40

50

RS RY TS TY

Cu

rem

oval

(m

g/g)

Cu-Zn

Fig. 5 (continued)

Water Air Soil Pollut

first hydrolysis constant was observed (Kinraide andYermiyahu 2007; Appel et al. 2008; Zhu et al. 2010).In this regard, metals with higher hydrolysis constant(eventually lower pKh) are specifically sorbed to agreater extent than that found for the present study(Vidal et al. 2009).

Lead removal capacity of RS clay sample clearlyincreased from 18.76 mg/g in Pb–Cd–Cu mixture to45.38 mg/g in Pb–Cd–Zn system. This indicates thehigh competitive behavior of Cu(II) toward Pb(II),probably because of their comparable relative bindingstrength, higher electronegativity and lower ionizationpotential (Kinraide and Yermiyahu 2007). Similar

results were obtained for RY, TS and TY samples, butthe removal increased to a much lower extent probablybecause of the removal of carbonate minerals, thepresence of which usually enhance the removaleffectiveness of Pb(II). In Pb–Cu–Zn and Pb–Cd–Cusystems, the studied clay samples removed similaramounts of Pb(II), but much lower than those recordedin Pb–Cd–Zn mixture. This was expected since thepresence of Cu(II) suppressed the removal of Pb(II),while Zn(II) and Cd(II) poorly affected the removal ofPb(II). Therefore, the removal of Pb(II) increased forall samples in Pb–Cd–Zn system, but decreased in Pb–Cu–Zn mixture indicating the antagonistic effect of Zn(II) and Cu(II). As for copper, the removal efficienciesin Pb–Cd–Cu and Pb–Cu–Zn systems were quitesimilar due to the comparable ionic properties of Zn(II) and Cd(II), which imply that the two ions wouldcompete similarly with Cu(II) ions. Moreover, thehighest removal of Cu(II) by the studied clay sampleswas recorded in Cd–Cu–Zn system because of the pooreffects of both Cd(II) and Zn(II) on Cu(II) removal, asmentioned earlier.

Finally, it is noteworthy that the studied claysamples removed low amounts of Zn(II) and Cd(II)compared with Pb(II) and Cu(II) in all cases probablybecause of their low electronegativity combined withhigh hydrolysis constant and ionization potential(Table 3). In addition, it was clearly observed thatboth RY and TY clay samples removed comparablehigher amounts of metals than RS and TS, except forPb–Cd–Zn system (Fig. 6). Furthermore, the removalamounts of both Pb(II) and Cu(II) by RS clay sampleclearly decreased after the removal of carbonateminerals. The solubility products of CuCO3 andPbCO3 were 1.4×10−10 and 7.4×10−14, respectively.This clearly indicated the precipitation of leadcarbonate and copper carbonate under the conditionof our experiments.

To summarize, the adsorption of Pb(II), Cu(II), Cd(II) and Zn(II) by natural clay samples depended onthe properties of both metals and adsorbents (Sdiri etal. 2011a). According to Usman (2008) and Vidal etal. (2009), the physical properties of the studiedmetals, including electronegativity, softness, ionicradius and hydrolysis constant may affect metalsadsorption. The correlation between metal propertiesand its removal capacity by natural clay samplesseemed to be even more important than the adsorbentproperties (Pagnanelli et al. 2003).

Table 2 Inhibitory effect of a metal on the adsorption ofcompeting metals

Sample Inhibited Inhibitor

Pb(II) Cd(II) Cu(II) Zn(II)

RS Pb(II) – – – –

Cd(II) 79.394 – 82.616 66.487

Cu(II) 55.910 13.197 – 10.179

Zn(II) 75.952 72.536 88.502 –

RY Pb(II) – 8.564 58.705 24.438

Cd(II) 60.021 – 68.771 48.948

Cu(II) 33.011 10.644 – 46.422

Zn(II) 50.343 20.643 62.602 –

TS Pb(II) – – 34.411 –

Cd(II) 72.557 – 77.836 62.478

Cu(II) 39.601 13.504 – 18.704

Zn(II) 66.112 43.778 72.222 –

TY Pb(II) – 2.653 46.128 22.096

Cd(II) 52.765 – 66.985 44.374

Cu(II) 34.743 10.721 – 25.626

Zn(II) 51.184 23.125 63.542 –

Table 3 Physical characteristics of the studied cations (Kinraideand Yermiyahu 2007)

Cation RBS PE IR IP HR pKh HI

Cd2+ 2.15 1.69 0.97 8.99 4.26 10.1 0.081

Cu2+ 2.66 1.90 0.73 7.73 4.19 7.5 0.104

Pb2+ 2.68 2.33 1.19 7.42 4.01 7.6 0.131

Zn2+ 2.20 1.65 0.74 9.39 4.3 9.0 0.115

RBS relative binding strengths, PE Pauling electronegativity, IRionic radius, IP ionization potential, HR hydrated radius, pKh,negative log of the fist hydrolysis constant, HI hardness index

Water Air Soil Pollut

4.2.4 Adsorption in Quadruple System

Mixed metal sorption of Pb(II), Cd(II), Cu(II) and Zn(II)ions by natural clay samples was also investigated toestablish the effect of the presence of three metal ions onthe removal of each one of them. In this analysis, thereaction was simulated by shaking with clay samples(raw and carbonate free samples) at 200 rpm for 60 minto reach equilibrium with the initial metal solution of pH6. The initial solution concentration was 60, 10, 40 and20 mg/l for Pb(II), Cd(II), Cu(II) and Zn(II), respectively,while the amount of clay was 1 g/l. Figure 7 reveals thatthe studied clay samples had a high affinity to Pb(II),leading to the highest removal amounts. For instance,Pb(II) removal efficiency was 15.9, 16.1, 8.3 and15.1 mg/g for RS, RY, TS and TY samples, respectively.The sorption capacity of Cu(II) was about 7.7, 8.1, 4.0and 7.7 mg/g for RS, RY, TS and TY samples,respectively. This also indicated the high affinity to Cu(II), though lower than Pb(II). According to theexperimental results, the removal of Pb(II), Cd(II), Cu(II) and Zn(II) by natural clay samples is governed bythe properties of the studied metals (i.e., relative bindingstrength, hydrated radius and hydrolysis constant) aswell as the adsorbent properties. Therefore, it wasexpected that Pb(II) and Cu(II) would show highcompetitive abilities for the surface sites. In addition,the precipitation of lead and copper carbonates may be

adopted as a possible removal mechanism due to thehigh carbonates content in the original sample (RS).

As for Cd(II) and Zn(II), it was clearly observed thatonly RY and TY clay samples removed measurableamounts of Zn(II), but negligible amounts of Cd(II).Thus, it could be assumed that the studied clay samplesremoved the following sequence Pb(II)>Cu(II)>Zn(II)>Cd(II), indicating that the studied metals competitivelyinteracted with original and treated clay samples.

4.3 Comparison to Other Studies

Based on previous relevant studies, the amount of heavymetals removed by various natural and syntheticadsorbents is highly variable (Table 4). In the current

0

10

20

30

40

50

RS RY TS TY

Pb

rem

oval

(m

g/g)

Pb-Cd-Cu

Pb-Cd-Zn

Pb-Cu-Zn

0

10

20

30

40

50

RS RY TS TY

Cd

rem

oval

(m

g/g)

Pb-Cd-Cu

Pb-Cd-Zn

Cd-Cu-Zn

0

10

20

30

40

50

RS RY TS TYZ

n re

mov

al (

mg/

g)

Pb-Cd-Zn

Pb-Cu-Zn

Cd-Cu-Zn

0

10

20

30

40

50

RS RY TS TY

Cu

rem

oval

(m

g/g)

Pb-Cd-Cu

Pb-Cu-Zn

Cd-Cu-Zn

Fig. 6 Adsorption of Pb(II), Cd(II), Cu(II) and Zn(II) by natural clays in ternary systems

15.9 16.1

8.3

15.1

7.7 8.1

4.0

7.7

0.52.9

0.82.9

0 0.8 0.10.8

0

10

20

RS RY TS TY

Met

al r

emov

al (

mg/

g)

Pb

Cu

Zn

Cd

Fig. 7 Adsorption of Pb(II), Cd(II), Cu(II) and Zn(II) bynatural clays in quadruple system

Water Air Soil Pollut

Table

4Com

parisonof

adsorptio

ncapacity

(mg/g)

with

thoseof

previous

remov

alstud

ieswith

naturalclays

Sorbent

BinarySystems

Pb–Cd

Ref.

Pb–Cu

Ref.

Pb–Zn

Ref.

Cd–Cu

Ref.

Cd–Zn

Ref.

Cu–Zn

Ref.

Pb

Cd

Pb

Cu

Pb

Zn

Cd

Cu

Cd

Zn

Cu

Zn

RS

48.04

1.15

15.63

8.62

50.10

3.66

0.97

16.97

1.87

4.18

17.56

1.75

RY

34.61

3.61

19.90

11.14

28.60

6.64

14.86

2.82

4.61

10.61

12.85

5

TS

24.87

1.57

11.97

6.62

21.14

3.05

9.48

1.27

2.15

5.06

8.91

2.5

TY

35.20

4.45

19.48

10.95

28.17

7.03

14.98

3.11

5.24

11.07

12.48

5.25

Bentonite

18.08

17.09

Morales

Futalan

etal.(2011)

Soil

6.69

3.61

App

elet

al.(200

8)Natural

Zeolite

21.74

2.6

13.93

1.2

4.64

1.11

Minceva

etal.(200

7)Sedim

ents

20.51

2.25

Oh et

al.(200

9)Hyd

roxy

apatite

95.96

76.52

Coram

iet

al.(200

7)

Calcareou

ssoil

0.8

0.3

Mesqu

itaandVieira

eSilv

a(199

6)

Bentonite

7.56

5.8

Ulm

anu

etal.(200

3)kaolinite

–4.36

Illite

104.2

Oub

agaranadin

etal.(201

0)

Ternarysystem

s

Pb–Cd–Cu

Ref.

Pb–Cd–Zn

Ref.

Pb–Cu–

Zn

Ref.

Cd–Cu–Zn

Ref.

Pb

Cd

Cu

Pb

Cd

Zn

Pb

Cu

Zn

Cd

Cu

Zn

RS

18.76

0.3

8.75

45.38

0.8

2.58

20.42

9.52

0.76

0.76

16.95

1.86

RY

23.11

1.53

10.80

29.15

2.18

621

.18

10.38

3.39

1.97

12.18

4.94

TS

13.42

0.37

5.43

22.45

0.85

2.6

13.10

6.34

1.25

0.83

7.50

2.32

TY

22.93

1.52

10.05

27.74

2.38

6.22

21.10

9.75

3.36

2.04

11.05

5.14

Smectite

16.76

11.23

11,86

Arpa

etal.

(200

0)Mon

tmorillon

ite2.49

1.46

0.89

Saha

etal.(200

2)Brownfielddeposits

0.58

0.24

60.28

Markiew

icz-

Patkow

ska

etal.

(2004)

Greek

lignite

9.95

9.87

9.82

9.95

9.75

9.82

9.87

9.75

9.82

Pentariet

al.(200

9)

Water Air Soil Pollut

study, the original clay samples demonstrated substan-tial removal amounts of metals when compared withtreated forms. It was clear from the study data thatthe removal efficiency was dependent upon thephysicochemical characteristics of the clay sampleand the metal removed. In single element system,the calculated Langmuir capacities were 131.58,51.28, 32.89 and 50.76 mg/g for lead, and 27.40,17.89, 12.97 and 17.10 mg/g for copper removal byRS, RY, TS and TY samples, respectively (Sdiri et al.2011a). The lowest removal occurred for cadmiumand zinc. This was further confirmed in binary,ternary and quadruple systems (Table 4). Forinstance, the removal of Pb(II) by RS sampledecreased to 48.04, 15.63 and 50.10 mg/g in Pb–Cd, Pb–Cu and Pb–Zn systems, respectively, whilethe removal amounts of Cu(II) decreased to 11.14,14.86 and 12.85 mg/g for Cu(II) adsorption by RYsample in Pb–Cu, Cd–Cu and Cu–Zn, respectively.This may indicate the competitive effects as well asthe high affinity of carbonate rich clay to both Cu(II)and Pb(II). This competitive behavior might beexplained by the difference in the physical propertiesof the studied metals. The adsorption of the metals inmulti-element system followed their correspondingbehavior in single-element systems. Our results revealedthat metals with higher hydrolysis constants (Pb(II)and Cu(II)) were more readily adsorbed to thesurface at lower pH. Similar results were found bySrivastava et al. (2005) who mentioned that Cu(II)and Pb(II) removal started at pH 5.0, while Cd(II)adsorbed onto permanent charge sites at pH 7.5.Moreover, just the presence of Cu(II) and Pb(II)suppressed the adsorption of Cd(II) and Zn(II) due totheir high relative binding strength, electronegativityand hydrolysis constant combined with lower ionizationpotential, and therefore higher competitive capacity(Table 3).

In the case of RY and TY, the removal remainedroughly constant, but decreased for the treated S sample(TS) with regards to RS because of carbonates removal.All these results indicate higher removal efficiency forthe present clay samples than was shown by multiplesorbents, especially for Pb(II) and Cu(II) (Table 4). Thisindicates that the clay samples of southern Tunisiaexhibited greater removal efficiency than thosereported in literature and confirm their appropriateuse for the removal Pb(II), Cd(II), Cu(II) and Zn(II) insingle and mixed systems.

5 Conclusions

The adsorption of heavy metal cations (Pb(II), Cd(II),Cu(II) and Zn(II)) in single and multi-elements systemonto clay samples of the Coniacian–Early CampanianAleg formation, Tunisia, followed Pb(II)>Cu(II)>Zn(II)>Cd(II). In the binary, ternary and quadruple solutesystems, the adsorption capacity decreased for eachmetal due to competitive effect. In addition, the removalof Pb(II) ions in the presence of other metals showedgreater potential than the removal of other metals in thepresence of Pb(II). Such findings are contingent uponsome physical properties of the studied metals (i.e.,relative binding strength, hydrated radius, electronega-tivity and hydrolysis constant). Our study showed thatgood adsorptive capacities could be achieved under theoperating conditions of 60 min contact time, pH 6 andclay concentration of 1 g/l at 25°C. The obtainedresults demonstrated that, in addition to the adsorp-tion to the reactive sites, ion exchange with Ca(II)and the precipitation as metal carbonates werepossibly involved as effective immobilization mecha-nisms, especially for Pb(II) and Cu(II). From thecomparative study of the current and previously pub-lished data, the studied clay samples exhibited greaterremoval efficiency than was reported in literature. It istherefore possible to affirm that the Late Cretaceous claysfrom Tunisia can be effectively used as natural adsorbentsfor toxic heavy metals removal from wastewaters.Finally, more detailed studies on the effects of environ-mental parameters, including the effect of temperatureand ionic strength may be helpful for the development ofa cost effective method for heavy metals removal inaqueous systems.

Acknowledgements The authors gratefully acknowledge thefinancial support provided by Borj Cedria Science andTechnology Park, Tunisia, in cooperation with the Japan Bankfor International Cooperation (JBIC). The help of Mrs. SeikoNemoto, assistant researcher in JIRCAS, Tsukuba, Mrs. HirokoSado and Mr Koichi Shimotori, research associate in theChemical Analysis Division, University of Tsukuba, is greatlyappreciated. The authors also extend their thanks to ProfessorAhmet Sari from Gaziosmanpaşa University, Turkey for hisprompt reviews and the time allocated to our manuscript.

References

Abdeljaouad, S. (1983). Etude stratigraphique, sédimentologiqueet structurale de l’extrémité orientale de la Chaîne Nord desChotts. PhD thesis, Fac Sci, Univ Tunis II, Tunisia.

Water Air Soil Pollut

Al-Degs, Y. S., El-Barghouthi, M. I., Issa, A. A., Khraisheh, M.A., & Walker, G. M. (2006). Sorption of Zn(II), Pb(II), andCo(II) using natural sorbents: equilibrium and kineticstudies. Water Research, 40, 2645–2658.

Appel, C., Ma, L. Q., Rhue, R. D., & Reve, W. (2008).Sequential sorption of lead and cadmium in three tropicalsoils. Environmental Pollution, 155, 132–140.

Arpa, C., Say, R., Satiroglu, N., Yurum, S., Bektas, Y., & Genc, O.(2000). Heavy metal removal from aquatic systems bynorthern Anatolian smectites. Turkish Journal of Chemistry,24, 209–215.

Ben Hadj, A., Jedoui, Y., Dali, T., Ben Salem, N., Memmi, L.(1985). Carte géologique de la Tunisie à 1/500,000, OfficeNational des Mines, Service Géologique National, Tunisie

Burollet, P. F. (1956). Contribution à l’étude stratigraphique de laTunisie Centrale.Annales deMines et de laGéologie, 18, 345.

Chaari, I., Fakhfakh, E., Chakroun, S., Bouzid, J., Boujelben, N.,Feki, M., et al. (2008). Lead removal from aqueous solutionsby a Tunisian smectitic clay. Journal of Hazardous Materials,156, 545–551.

Corami, A., Mignardi, S., & Ferrini, V. (2007). Copper and zincdecontamination from single- and binary-metal solutionsusing hydroxyapatite. Journal of Hazardous Materials,146, 164–170.

Darder, M., Colilla, M., & Ruiz-Hitzky, E. (2005). Chitosan–claynanocomposites: application as electrochemical sensors.Applied Clay Sciences, 28, 199–208.

Echeverria, J. C., Morera, M. T., Mazkiaran, C., & Garrido, J. J.(1998). Competitive sorption of heavy metal by soils.Isotherms and fractional factorial experiments. EnvironmentalPollution, 101, 275–284.

Eloussaief, M., & Benzina, M. (2010). Efficiency of natural andacid-activated clays in the removal of Pb(II) from aqueoussolutions. Journal of Hazardous Materials, 178, 753–757.

Eloussaief, M., Jarraya, I., & Benzina, M. (2009). Adsorption ofcopper ions on two clays from Tunisia: pH and temperatureeffects. Applied Clay Sciences, 46, 409–413.

Eloussaief, M., Kallel, N., Yaacoubi, A., & Benzina, M. (2011).Mineralogical identification, spectroscopic characterization,and potential environmental use of natural clay materialson chromate removal from aqueous solutions. ChemicalEngineering Journal, 168, 1024–1031.

Ghorbel-Abid, I., Galai, K., & Trabelsi-Ayadi, M. (2010).Retention of chromium (III) and cadmium (II) fromaqueous solution by illitic clay as a low-cost adsorbent.Desalination, 256, 190–195.

Guerra, D. L., Bastita, A. C., Corrêa da costa, P. C., Viana, R. R., &Airoldi, C. (2010). Adsorption of arsenic ions on Braziliansepiolite: effect of contact time, pH, concentration, andcalorimetric investigation. Journal of Colloid and InterfaceScience, 346, 178–187.

Jarraya, I., Fourmentin, S., Benzina, M., & Bouaziz, S. (2010).VOCs adsorption on raw and modified clay materials.Chemical Geology, 275, 1–8.

Kaya, A., & Hakan Ören, A. (2005). Adsorption of zinc fromaqueous solutions to bentonite. Journal of HazardousMaterials, 125, 183–189.

Kinraide, T. B., & Yermiyahu, U. (2007). A scale of metal ionbinding strengths correlating with ionic charge, Paulingelectronegativity, toxicity, and other physiological effects.Journal of Inorganic Biochemistry, 101, 1201–1213.

Lin, S. H., & Juang, R. S. (2002). Heavy metal removal fromwater by sorption using surfactant-modified montmorillonite.Journal of Hazardous Materials, 92, 315–326.

Mahamadi, C., & Nharingo, T. (2010). Competitive adsorptionof Pb2+, Cd2+ and Zn2+ ions onto Eichhornia crassipes inbinary and ternary systems. Bioresource Technology, 101,859–864.

Markiewicz-Patkowska, J., Hursthouse, A., & Przybyla-Kij, H.(2004). The interaction of heavy metals with urban soils:sorption behaviour of Cd, Cu, Cr, Pb and Zn with a typicalmixed brownfield deposit. Environmental Pollution, 31,513–521.

Mejri, F., Burollet, P. F., Ben Ferjani, A. (2006). Petroleumgeology of Tunisia, a renewed synthesis. Mémoire N° 22,Entreprise tunisienne des activités petrolières, Tunisia.

Mesquita,M. E., &Vieira e Silva, J.M. (1996). Zinc adsorption by acalcareous soil. Copper interaction. Geoderma, 69, 137–146.

Minceva, M., Markovska, L., & Meshko, V. (2007). Removal ofZn2+, Cd2+ and Pb2+ from binary aqueous solution bynatural zeolite and granulated activated carbon.MacedonianJournal of Chemistry and Chemical Engineering, 26, 125–134.

Morales Futalan, C., Kan, C., Lourdes Dalida, M., Hsien, K.,Pascua, C., & Wan, M. (2011). Comparative and competitiveadsorption of copper, lead, and nickel using chitosanimmobilized on bentonite. Carbohydrate Polymers, 83,528–536.

Oh, S., Kwak, M. Y., & Shin, W. S. (2009). Competitivesorption of lead and cadmium onto sediments. ChemicalEngineering Journal, 152, 376–388.

Oubagaranadin, J. U. K., Murthy, Z. V. P., & Mallapur, Z. V. P.(2010). Removal of Cu(II) and Zn(II) from industrialwastewater by acid-activated montmorillonite–illite typeof clay. Comptes Rendus Chimie, 13, 1359–1363.

Pagnanelli, F., Esposito, A., Toro, L., & Veglio, F. (2003). Metalspeciation and pH effect on Pb, Cu, Zn and Cd biosorptiononto Sphaerotilus natans: Langmuir-type empirical model.Water Research, 37, 627–633.

Pentari, D., Perdikatsis, V., Katsimicha, D., & Kanaki, A.(2009). Sorption properties of low calorific value Greeklignites: removal of lead, cadmium, zinc and copper ionsfrom aqueous solutions. Journal of Hazardous Materials,168, 1017–1021.

Qin, F., Wen, B., Shan, X. Q., Xie, Y. N., Liu, T., Zhang, S. Z., et al.(2006). Mechanisms of competitive adsorption of Pb, Cu,and Cd on peat. Environmental Pollution, 144, 669–680.

Saha, U. K., Taniguchi, S., & Sakurai, K. (2002). Simultaneousadsorption of cadmium, zinc, and lead on hydroxyaluminum-and hydroxyaluminosilicate–montmorillonite complexes.Soil Science Society of America Journal, 66, 117–128.

Sari, A., Tuzen, M., Citak, D., & Soylak,M. (2007a). Equilibrium,kinetic and thermodynamic studies of adsorption of Pb(II)from aqueous solution onto Turkish kaolinite clay. Journal ofHazardous Materials, 149, 283–291.

Sari, A., Tuzen, M., & Soylak, M. (2007b). Adsorption of Pb(II) and Cr(III) from aqueous solution on Celtik clay.Journal of Hazardous Materials, 144, 41–46.

Schaller, M. S., Koretsky, C. M., Lund, T. J., & Landry, C. J.(2009). Surface complexation modeling of Cd(II) adsorptionon mixtures of hydrous ferric oxide, quartz and kaolinite.Journal of Colloid and Interface Science, 339, 302–309.

Water Air Soil Pollut

Sdiri, A., Higashi, T., Bouaziz, S., & Benzina, M. (2010a).Synthesis and characterization of silica gel from siliceoussands of southern Tunisia. Arabian Journal of Chemistry.doi:10.1016/j.arabjc.2010.11.007.

Sdiri, A., Higashi, T., Hatta, T., Jamoussi, F., & Tase, N.(2010b). Mineralogical and spectroscopic characterization,and potential environmental use of limestone from theAbiod formation, Tunisia. Environmental Earth Sciences,61, 1275–1287.

Sdiri, A., Higashi, T., Hatta, T., Jamousssi, F., & Tase, N. (2011a).Evaluating the adsorptive capacity of montmorillonitic andcalcareous clays on the removal of several heavy metalsin aqueous systems. Chemical Engineering Journal, 172,37–46.

Sdiri, A., Higashi, T., Jamoussi, F., & Bouaziz, S. (2011b). Effectsof impurities on the removal of heavy metals by naturallimestones in aqueous systems. Journal of EnvironmentalManagement. doi:10.1016/j.jenvman.2011.08.002.

Srivastava, P., Singh, B., & Angove, M. (2005). Competitiveadsorption behavior of heavy metals on kaolinite. Journalof Colloid and Interface Science, 290, 28–38.

Ulmanu, M., Marañón, E., Fernández, Y., Castrillón, L., Anger,I., & Dumitriu, D. (2003). Removal of copper andcadmium ions from diluted aqueous solutions by low costand waste material adsorbents. Water, Air, and SoilPollution, 142, 357–373.

Usman, A. A. R. (2008). The relative adsorption selectivities ofPb, Cu, Zn, Cd and Ni by soils developed on shale in NewValley, Egypt. Geoderma, 144, 334–343.

Vidal, M., Santos, M. J., Abrão, T., Rodríguez, J., & Rigol, A.(2009). Modeling competitive metal sorption in a mineralsoil. Geoderma, 149, 189–198.

Zhu, J., Pigna, M., Cozzolino, V., Caporale, A. G., & Violante,A. (2010). Competitive sorption of copper(II), chromium(III) and lead(II) on ferrihydrite and two organomineralcomplexes. Geoderma, 159, 409–416.

Water Air Soil Pollut