Cn

TYa

b

Cc

Pd

h

•••

a

ARRAA

KEPCSE

1

tapf

Yf

0h

Journal of Hazardous Materials 262 (2013) 464– 471

Contents lists available at ScienceDirect

Journal of Hazardous Materials

jou rn al hom epage: www.elsev ier .com/ locate / jhazmat

helant extraction of heavy metals from contaminated soils usingew selective EDTA derivatives

ao Zhanga, Jun-Min Liub, Xiong-Fei Huanga, Bing Xiaa, Cheng-Yong Sub, Guo-Fan Luoa,ao-Wei Xub, Ying-Xin Wua, Zong-Wan Maoc, Rong-Liang Qiua,d,∗

School of Environmental Science and Engineering, Sun Yat-sen University, Guangzhou 510275, PR ChinaMOE Key Laboratory of Bioinorganic and Synthetic Chemistry, State Key Laboratory of Optoelectronic Materials and Technologies, School of Chemistry andhemical Engineering, Sun Yat-Sen University, Guangzhou 510275, PR ChinaMOE Key Laboratory of Bioinorganic and Synthetic Chemistry, School of Chemistry and Chemical Engineering, Sun Yat-Sen University, Guangzhou 510275,R ChinaGuangdong Provincial Key Laboratory of Environmental Pollution Control and Remediation Technology, Guangzhou 510275, PR China

i g h l i g h t s

Two EDTA derivatives were synthesized to enhance the selectivity of chelators.PDTA had the highest stability constants for Cu2+ and Ni2+.PDTA had the highest overall selectivity for trace metals over major cations.

r t i c l e i n f o

rticle history:eceived 13 April 2013eceived in revised form 30 July 2013ccepted 26 August 2013vailable online 1 September 2013

eywords:DTA derivativesotentiometric titrationu(II)

a b s t r a c t

Soil washing is one of the few permanent treatment alternatives for removing metal contaminants.Ethylenediaminetetraacetic acid (EDTA) and its salts can substantially increase heavy metal removalfrom contaminated soils and have been extensively studied for soil washing. However, EDTA has a poorutilization ratio due to its low selectivity resulting from the competition between soil major cations andtrace metal ions for chelation. The present study evaluated the potential for soil washing using EDTA andthree of its derivatives: CDTA (trans-1,2-cyclohexanediaminetetraacetic acid), BDTA (benzyldiaminete-traacetic acid), and PDTA (phenyldiaminetetraacetic acid), which contain a cylcohexane ring, a benzylgroup, and a phenyl group, respectively. Titration results showed that PDTA had the highest stabilityconstants for Cu2+ and Ni2+ and the highest overall selectivity for trace metals over major cations. Equi-

electivityxtraction

librium batch experiments were conducted to evaluate the efficacy of the EDTA derivatives at extractingCu2+, Zn2+, Ni2+, Pb2+, Ca2+, and Fe3+ from a contaminated soil. At pH 7.0, PDTA extracted 1.5 times moreCu2+ than did EDTA, but only 75% as much Ca2+. Although CDTA was a strong chelator of heavy metal ions,its overall selectivity was lower and comparable to that of EDTA. BDTA was the least effective extractantbecause its stability constants with heavy metals were low. PDTA is potentially a practical washing agent

ith tra

for soils contaminated w

. Introduction

Heavy metal contamination of soils, resulting from rapid indus-rialization, increased urbanization, modern agricultural practices,

nd inappropriate waste disposal methods, has become a seriousroblem worldwide [1–3]. The available remediation technologiesor heavy metal-contaminated soils are mainly divided into two

∗ Corresponding author at: School of Environmental Science and Engineering, Sunat-sen University, Guangzhou 510275, PR China. Tel.: +86 20 84113454;

ax: +86 20 84113616.E-mail address: eesqrl@mail.sysu.edu.cn (R.-L. Qiu).

304-3894/$ – see front matter © 2013 Elsevier B.V. All rights reserved.ttp://dx.doi.org/10.1016/j.jhazmat.2013.08.069

ce metals© 2013 Elsevier B.V. All rights reserved.

groups: namely immobilization, such as in situ chemical fixation,and separation, such as soil washing. Chelant-enhanced soil wash-ing is a technology that is potentially useful for the economicallyfeasible remediation of contaminated soils [4–8].

The chelating agent ethylenediaminetetraacetic acid (EDTA) andits salts have been extensively studied for their potential use in soilwashing [9–12]. They have been reported to appreciably increasethe dissolution and mobilization of cationic heavy metals [13,14].EDTA has low biodegradability in soil and a high efficiency of

metal extraction through the formation of thermodynamically sta-ble and soluble metal–EDTA complexes [15,16]. In addition, recentadvances in recovery and recycling techniques of used EDTA haveenhanced its appeal [17].

T. Zhang et al. / Journal of Hazardous Materials 262 (2013) 464– 471 465

Table 1Soil properties.

−1 Clay −1 −1 −1 −1

10

dmo(meCedcsetifae

sEoidabatapdot

2

2

srT

wsdtmu

Soil pH OM (%) CEC (cmol kg ) Sand (%) Silt (%)

GD 7.3 5.1 9.5 61 31

A majority of the literature focused on demonstrating the reme-iation capabilities of EDTA has found that extraction of heavyetals was faster and more complete with increased quantities

f added EDTA. Competition between the major cations of the soile.g., Ca2+, Mg2+, and Fe3+) and minor cations for chelation by EDTA

ay be one of the factors affecting the efficiency of trace metalxtraction [18–23]. As illustrated in previous studies [24,25], whena solubility in calcareous soils is raised, the effectiveness of EDTAxtraction is diminished significantly, increasing the cost of reme-iation. For non-calcareous soils, Fe and Al dissolution may be morerucial, in view of their high tendency for complexation (i.e., largetability constants). Excessive addition of chelating agent can causextensive dissolution of soil minerals and organic matter, leadingo alteration of soil physical and chemical properties and even dis-ntegration of soil structure, which could render the soil unfit foruture use for vegetation or construction. Therefore, there has been

growing need to develop highly selective chelating agents for thextraction of heavy metal ions from polluted soils.

In previous studies, EDTA has been modified to improve itselectivity in chelating target metal ions [26–28]. Highly selectiveDTA derivatives have a wide range of application in the fieldsf analytical chemistry, biology and medicine, as well as in manyndustrial processes. In this study, two new EDTA derivatives wereesigned and synthesized with the goal of enhancing its selectivitys a chelating agent. These EDTA derivatives contain a phenyl orenzyl group directly bonded to the nitrogens of the ethylenedi-mino group, and thus potentially are more sterically constrainedhan the parent compound. The objective of this study was to char-cterize these EDTA derivatives in aqueous solution and assess theirotential as selective washing agents. Batch experiments were con-ucted to determine their efficacy in the simultaneous extractionf trace metal ions and major cations from contaminated soils ando investigate the extraction mechanisms.

. Materials and methods

.1. Soil Characteristics

Soil samples were collected from 0.7 to 1.7 m below the groundurface at a demolished industrial site in South China, air-dried atoom temperature (20–30 ◦C), and passed through a 2 mm sieve.he soil properties in Table 1 were the average of three replicates.

Various soil physical and geochemical characterization testsere carried out. The physical and chemical characteristics of the

oil are shown in Table 1. The metal concentrations in soil were

etermined by acid digestion with HF–HClO4–HNO3 and induc-ively coupled plasma optical emission spectrometry (ICP-OES)

easurement (5300DV, Perkin Elmer). Soil pH was determinedsing a 1:5 soil-to-water ratio and pH meter. The CEC of the sample

Fig. 1. Synthesis

(%) Cu (mg kg ) Ni (mg kg ) Pb (mg kg ) Zn (mg kg )

1388 26 507 236

was determined by the ammonium acetate/sodium acetate method[29]. Organic matter content was determined by heating the driedsamples at 350 ◦C for 5 h [30]. The particle size distribution wasestablished by mechanical sieving, followed by the hydrometermethod [31].

2.2. Synthesis of EDTA derivatives

Phenyldiaminetetraacetic acid (PDTA) was synthesized as illus-trated in Fig. 1 according to a previously published procedure[32]. A solution of 1,2-diaminobenzene (1.1 g, 10 mmol), ethyl bro-moacetate (6.7 mL, 60 mmol), sodium iodide (1.3 g, 8.5 mmol), anddiisopropylethylamine (8.3 mL 50 mmol) in 10 mL acetonitrile wasrefluxed under nitrogen for 7 h, then cooled and poured into 50 mLwater. The resulting mixture was extracted with dichloromethane(3× 40 mL). The extract was dried and was concentrated to givebrown oil, which was purified by flash chromatography usingpetroleum ether/ethylacetate (10:1, v/v) as eluant. The productobtained was 2.1 g (45%) of a white solid (A) (Fig. 1); a mixtureof ethanol, complex A, and NaOH (14.0 M) were refluxed at 78 ◦Cfor 24 h. The solution was cooled to room temperature, ethanol wasevaporated, and the residue was washed with 6.0 M HCl solutionand dried under high vacuum to give a white solid (B) (Fig. 1).

Benzyldiaminetetraacetic acid (BDTA) was synthesized as illus-trated in Fig. 2. To a mixture of 1,2-bis(bromomethyl)benzene(1.0 mmol) in 5.0 mL dry CH3CN was added diethyliminodiacetate(2.0 mmol) and anhydrous K2CO3 (10 mmol). The resulting reac-tion mixture was stirred at room temperature for 5 h, then thesolids were filtered off and solvent was removed under vacuum.The residue was again dissolved in ethyl acetate and washed withwater and saturated NaOH solution, which was purified by flashchromatography using petroleum ether/ethylacetate (10:1, v/v) aseluant. Light yellowish oil (A) (Fig. 2) was obtained by evaporationof the organic phase. Then the residue was dissolved in the solu-tion of KOH. The mixture was stirred at room temperature for 24 h,and then resulting mixture was poured into ice water, after whichit was acidified with concentrated HCl (pH ∼ 1) to obtain a whitesolid (B) (Fig. 2).

2.3. Potentiometric titration

An automatic titrator (Metrohm 702GPD Titrino) coupled to aMetrohm electrode was used and calibrated according to the Granmethod [33,34]. The electrode system was calibrated with buffersand checked by titration of HClO4 with NaOH solution(0.01 M).

The thermostated cell contained 25 mL of 1.0 mM species (free lig-and/metal ligand complex) in aqueous solution with ionic strengthmaintained at 0.10 M by potassium nitrate. All titrations were car-ried out on the aqueous solutions under nitrogen at 298 ± 0.1 K, and

of PDTA.

466 T. Zhang et al. / Journal of Hazardous Materials 262 (2013) 464– 471

thesis

iipt2[

2

meow(atatv

doclp6t(4c

2

MW3(at

3

3

dsvteo(u

bonded to nitrogens of an ethylenediamino group [32].In total, three deprotonation events were observed for the

free ligands PDTA and BDTA (Fig. 3 and Table 2). At pH < 4.0, the

Fig. 2. Syn

nitiated by adding fixed volumes of 0.10 M standard NaOH in smallncrements to the titrated solution. Triplicate measurements wereerformed, for which the experimental error was below 1%. Theitration data were fitted from the raw data with the Hyperquad000 program to calculate the log ̌ and the pKa values of species35–38].

.4. Batch experiments

In the batch experiments, contaminated soil was mixed with aeasured volume of chelating agent solution in 50-mL polyethyl-

ne tubes. To probe the influence of EDTA derivative concentrationn metal mobility in the washed soil, every 1.000 g of soil was mixedith 20 mL chelating agent solution of different concentrations

0.005 M, 0.01 M, 0.02 M, 0.05 M and 0.1 M) in tubes. The tubes weregitated using a thermostatic shaker at 180 rpm at room tempera-ure (25 ± 2 ◦C) for a given time. The suspensions were centrifugedt 5000 rpm for 10 min and the supernatants were then filteredhrough a 0.45 �m membrane. All solutions were stored in amberials at 4 ◦C prior to analysis.

Another two sets of batch equilibrium experiments were con-ucted to study the influence of solution pH and liquid-to-soil ration metal extraction by the EDTA derivatives. An initial chelatoroncentration of 0.02 M was used in both sets. In the first set, theiquid-to-soil ratio was 20:1 (i.e., 20 mL g−1 soil) while the solutionH were adjusted using NaOH or HNO3 to pH values of 3.0, 4.0, 5.0,.0, 7.0, 8.0, 9.0, 10.0, 11.0, and 12.0. In the second set, the solu-ion pH was kept at pH 6.0 while five liquid-to-soil ratios of 2.5:12.5 mL g−1), 5:1 (5 mL g−1), 10:1 (10 mL g−1), 20:1 (20 mL g−1), and0:1 (40 mL g−1) were studied. These ratios were achieved using aonstant solution soil mass of 1.0 g.

.5. Analytical methods

The 1H NMR spectra of EDTA derivatives was obtained byercury-Plus 300 1H Nuclear Magnetic Resonance (VARIAN, USA).e determined the molecular weight of PDTA and BDTA as 339 and

64, respectively, with electro spray ionization mass spectroscopynegative ion mode). The heavy metals of concern (Cu, Ni, Zn and Pb)nd soil component elements (Ca and Fe) were measured by induc-ively coupled plasma optical emission spectrometry (ICP-OES).

. Results and discussion

.1. Acid–base properties of EDTA derivatives

The protonation constants of PDTA and BDTA in water wereetermined with a view to assessing their acid–base properties,ince these properties control what species exist in solution atarious pH values. Both PDTA and BDTA have six potential siteshat can bind with a proton, including the two nitrogens of the

thylenediamino groups and the four carboxyl groups. However,nly three deprotonation events were observed by potentiometryTable 2). For comparison, Table 2 also includes the literature val-es and determined values for EDTA. The agreement of the obtained

of BDTA.

pKa values with the protonation constants of EDTA (2.67, 6.16, and10.26) is acceptable.

The pKa1 value reflects the acidity of the carboxyl groups, andthe pKa2 and pKa3 values characterize the basicity of the aminonitrogens. The pKa2 and pKa3 values of CDTA and BDTA were similarto those of EDTA. The pKa3 value of CDTA was higher as a result ofthe electron-donating effect of the cyclohexane ring attached to theamino nitrogens.

The obtained pKa1 value of PDTA (2.79 ± 0.01), correspondingto the protonation of carboxyl groups, was lower than that ofEDTA (3.05 ± 0.02). The lower pKa1 value of PDTA indicates that theacidity of the carboxyl groups was increased due to the electron-withdrawing effect of the aromatic group.

Similarly, PDTA had lower pKa2 and pKa3 values in comparisonwith EDTA, indicating that the basicity of the nitrogen atoms issignificantly lower for PDTA relative to EDTA [39]. This may beattributed to the lower electronic density of the nitrogens in theformer due to the presence of a phenyl group, since the electron-withdrawing effect is increased when an aromatic group is directly

Fig. 3. Distribution plots of species of the free ligands PDTA (1.0 mM) and BDTA(1.0 mM).

T. Zhang et al. / Journal of Hazardous Materials 262 (2013) 464– 471 467

Table 2Deprotonation constants (pKa) of EDTA, CDTA, PDTA, and BDTA.

Equilibrium EDTAa CDTAa EDTAb PDTAb BDTAb

H3Y−/H2Y2− (pKa1) 2.67 3.53 3.05 ± 0.02 2.79 ± 0.01 3.56 ± 0.03H2Y2−/HY3− (pKa2) 6.16 6.15 6.42 ± 0.02 3.70 ± 0.02 8.31 ± 0.05

3− 4− 10.50 ± 0.03 5.95 ± 0.03 10.00 ± 0.06

pPSwz

3

e

M

rwrid(

Table 3Equilibrium constants (log KML) of EDTA and derivatives in aqueous solution.

Metals EDTAa CDTAa PDTAb

Pb(II) 17.88 20.24 20.30 ± 0.05Cu(II) 18.70 21.92 24.84 ± 0.07Cd(II) 16.36 19.84 18.18 ± 0.03Zn(II) 16.44 19.35 17.99 ± 0.09Ni(II) 18.52 20.20 23.28 ± 0.05Hg(II) 21.50 24.79 19.73 ± 0.09Ca(II) 10.61 13.15 11.25 ± 0.05Mg(II) 8.83 11.07 10.41 ± 0.08

HY /Y (pKa3) 10.26 12.40

a Literature values.b Determined values.

ercentage of species H3Y−/H2Y2− was higher for BDTA than forDTA, which is consistent with the pKa2 values of BDTA and PDTA.imilarly the percentage of BDTA existing as the species HY3− (i.e.,ith one nitrogen deprotonated) at pH < 5.0 was approximately

ero, in accordance with the high pKa3 values of BDTA.

.2. Selectivity of EDTA derivatives

The metal–ligand (ML) complexation equilibrium constant, KML,xpresses the ligand’s affinity toward the target metal (reaction (I)).

2+ + H2Y2− KML−→2H+ + MY2− (I)

The equilibrium constants of PDTA and BDTA complexes with aange of metals were determined by pH potentiometric titration,ith the calculated results for the [M–PDTA] complex summa-

ized in Table 3. The [M–BDTA] complex, however, did not existn distribution plots of species, except for Cd–BDTA(OH), whichemonstrated that M–BDTA was not thermodynamically stableTable 4). We deduced that the structure of BDTA allowed it to form

Fig. 4. Extraction efficiency of different concentrations of chelating ag

a Literature values.b Determined values.

a seven-membered ring complex with most metal ions, but not astable chelate compound with a five-membered or six-memberedring.

Table 3 also shows that PDTA has higher selective capabilityfor Cu2+ and Ni2+, relative to Cd2+, Zn2+, Ca2+, and Mg2+, than dothe conventional chelators EDTA and CDTA. The differences in thelog KML values for Cu2+ and Ni2+ complexes with PDTA, CDTA, and

ents (liquid/solid = 20; contact time, 24 h; temperature, 25 ◦C).

468 T. Zhang et al. / Journal of Hazardous Materials 262 (2013) 464– 471

Table 4Equilibrium constants of M–BDTA in aqueous solution at 298 ± 0.1 K.

Species Equilibrium constants

log ̌ log KML

CdL 7.41 ± 0.08 –CdL(OH) 4.67 ± 0.07 3.57CaL(OH) 0.35 ± 0.02 –

MgL(OH) 0.30 ± 0.05 –ZnL(OH) 1.28 ± 0.04 –NiL(OH) 2.27 ± 0.04 –PbL(OH) 1.89 ± 0.02 –

ErpP

cga

p

p

Table 5pKtarget and SR values of EDTA and EDTA derivatives.

Chelating agents pKtarget SR

HgL(OH) −0.90 ± 0.05 –CuL(OH) – –

DTA are related to the steric constraints created by the benzeneing in PDTA and the cyclohexane ring in CDTA [40] that are notresent in EDTA. Because the radii of Ni2+ and Cu2+ are similar,DTA does not exhibit good selectivity between the two [41].

In evaluating EDTA derivatives, we determined the equilibriumonstants (log KML) of PDTA for complexation with six divalent tar-et metals, including Pb, Cu, Cd, Zn, Ni, and Hg [42], and with twombient divalent cations, Ca and Mg.

Ktarget = pKPb + pKCu + pKCd + pKZn + pKNi + pKHg

6(1)

Kambient = pKCa + pKMg

2(2)

Fig. 5. Extraction efficiency using chelator solutions with differen

EDTA 18.23 1.88CDTA 21.06 1.74PDTA 20.72 2.11

The selectivity of chelating agents toward heavy metals can bequantitatively computed on the basis of the selectivity ratio (SR),which is defined as:

SR = pKtarget

pKambient(3)

According to formula (3), a strong chelator will have largepKtarget (>12) and pKambient values, whereas a strong and selec-tive chelator toward heavy metals will have a large pKtarget anda relatively small pKambient, and thus a large SR value (>2) [42].The pKtarget and SR were calculated and summarized in Table 5,which may serve to guide in the selection of chelators that are botheffective and selective toward the six target metals. The resultsdemonstrated that CDTA and PDTA were strong chelators, eachhaving as large a pKtarget (>12) as EDTA. More interestingly, PDTA,compared with EDTA and CDTA, had a larger SR value (>2), showingthat it has a higher selectivity toward heavy metals.

3.3. Comparison of heavy metal removal with different chelators

3.3.1. Influence of concentrationEDTA and its derivatives in a range of concentrations were used

to extract the heavy metals Cu, Ni, Pb, and Zn from a contaminated

t liquid/soil ratios (contact time, 24 h; temperature, 25 ◦C).

T. Zhang et al. / Journal of Hazardous Materials 262 (2013) 464– 471 469

ncent

sitst5e7tcftcfehc

tcpEtFtPfttec

CDTA, increasing the liquid-to-soil ratio over this range generallyincreased the efficiency of extraction of Cu (from 21% to 54% byEDTA and from 24% to 61% by CDTA), Ni (from 0 to 12% by EDTAand from 0% to 10% by CDTA), Pb (from 8% to 18% by EDTA and from

Fig. 6. Influence of pH on extraction of metals using chelators (initial co

oil. The extent of removal of the four metals increased withncreasing concentration of the chelators from 0.005 to 0.10 M, buthe rate of increase for most metal–chelator combinations was lessteep above a chelator concentration of 0.02 M (Fig. 4). At a chela-or concentration of 0.02 M, the extraction of Cu was 62% by PDTA,9% by CDTA, 48% by EDTA, and 8.6% by BDTA, while at the high-st chelator concentration (0.10 M), these values increased to only3%, 65%, 60%, and 10%, respectively. Over the entire concentra-ion range, addition of PDTA caused the extraction increase slightlyompared with EDTA due to the low concentration of Ni in the soil,rom 4.4% to 15% by PDTA and from 2.2% to 11% by EDTA, respec-ively. Pb extraction by PDTA increased from 2.0% to 10%, while theorresponding Pb values for EDTA were much higher, increasingrom 8.1% to 21%. Zn showed a trend similar to that of Pb, withxtraction values reaching 13% for PDTA and 28% for EDTA at theighest chelator concentration. In light of these results, a chelatoroncentration of 0.02 M was selected for further studies.

As the chelator concentration was increased, only a small frac-ion of the added chelator was incorporated into metal–chelantomplexes, while the excess remained in the free form or was com-lexed with other cations (Ca, Mg, Fe, Al, etc.) [43]. Similarly, whenDTA is used for soil washing, not all EDTA added to soil bindshe target metals, because other ions in the soil such as Ca ande also interact with EDTA due to their high concentrations andhe relatively high stability of their complexes [44]. Above 0.02 M,DTA was the most efficient of the tested chelators at extracting Curom soil. This result can be attributed not only to stronger chela-

ion of Cu by PDTA compared to the other chelators, but also tohe weaker chelation between PDTA and other soil cations (Ca, Fe,tc.) and heavy metals, as evidenced by the pKML constants of theiromplexes (Table 3). In fact, PDTA had the second lowest Pb and

ration, 0.02 M; liquid/solid = 20; contact time, 24 h; temperature, 25 ◦C).

Zn extraction efficiencies, only slightly higher than those of BDTA,which is unable to form a stable chelate compound with a five- orsix-membered ring.

3.3.2. Influence of liquid-to-soil ratioAt a constant chelator concentration of 0.4 mmol g−1 soil, the

extraction efficiency of the target metals was measured as theliquid-to-soil ratio increased from 2.5:1 to 40:1. For EDTA and

Fig. 7. Distribution plots of species of 1.0 mM Cu:PDTA as a function of pH at 0.10 MKNO3 and (298 ± 0.1) K.

470 T. Zhang et al. / Journal of Hazardous Materials 262 (2013) 464– 471

F ation, ◦

l

62el

eiasthe

3

atn

iisMbHtabthItmCcitebtsatdi

ig. 8. Solubilization of Ca and Fe by EDTA and EDTA derivatives (a: initial concentriquid/solid = 20; contact time, 24 h; temperature, 25 ◦C).

% to 17% by CDTA), and Zn (from 8% to 17% by EDTA and from 6% to4% by CDTA), respectively (Fig. 5). However, for PDTA and BDTA,xtraction of the four metals was found to be independent of theiquid-to-soil ratio.

The extraction efficiency, as determined by the mole ratio ofxtracted potentially toxic trace metals to chelant inputs [45],ncreased with increasing liquid-to-soil ratio because a largermount of organic matter could be dissolved from a greater mass ofoil [8,43]. In addition, Cu has a high affinity toward organic mat-er, and the hydrophobic aromatic groups present in PDTA mayave facilitated soil organic matter dissolution, making Cu quicklyxtractable and independent of the liquid:soil ratio.

.3.3. Influence of solution pHAs shown in Fig. 6, extraction efficiency was highest at pH 3.0,

nd generally decreased as solution pH was increased from pH 3.0o pH 12.0. At pH 12.0, 24% of Cu was removed by PDTA but almostone was removed by EDTA and BDTA.

The important mobilization processes of heavy metals in soilnclude: (1) acidification, (2) competitive adsorption of other metalons or anions, (3) reductive and non-reductive dissolution of theolid phase, and (4) complexation of metal ions by ligands [46].etals bound to soil hydrous oxides can often be retrieved simply

y lowering pH because protons can promote oxide dissolution.ydrogen ions are also weak competitive cations which can replace

he adsorbed heavy metals via a cation exchange mechanism, ands the H+ ion concentration increases, the particle surface generallyecomes increasingly protonated and acquires a positive charge,hus promoting desorption of metals [13]. Therefore this is whyeavy metal extraction is higher at lower values of solution pH.

n addition, the pH-potentiometric titration results showed thathe equilibrium species of Cu at a 1:1 metal:PDTA ratio are pri-

arily ML, ML(OH), and ML(OH)2 (Fig. 7). The concentration of freeu2+ was very low in aqueous solution, and Cu is almost completelyomplexed above pH 3.5. As pH is increased, one water molecules deprotonated but remains in the inner coordination sphere ofhe complexed Cu2+, forming a stable chelate compound, whichxplains why even at a solution pH of 12.0, 24% of Cu was removedy PDTA. In practice, however, the pH is normally controlled withinhe range of 5.0–9.0 in order to eliminate adverse changes in theoil’s chemical and physical structure brought about by too acidic or

lkaline a solution [14]. As pH increased from 5.0 to 9.0, the extrac-ion efficiency of Cu by PDTA decreased from 72% to 56%, but theecrease was from 52% to 35% for EDTA. Thus, at constant pH, PDTA

s more efficient at extracting Cu from soil than is EDTA.

0.02 M; liquid/solid = 20; contact time, 24 h; temperature, 25 C. b: initial pH of 6.0;

3.3.4. Solubilization of major soil mineral elementsThe solubilization of soil mineral elements (i.e., Ca, and Fe) by

EDTA and its derivatives was also measured (Fig. 8). EDTA solubi-lized a large amount of Ca from the soil, with values that decreasedfrom 13 400 mg kg−1 to 4510 mg kg−1 as pH rose from pH 4.5 to pH9.0, while the amount of Ca solubilized by PDTA was considerablylower (10 500 mg kg−1 to 2900 mg kg−1 over the same pH range).Similarly, at pH 6.0 and a chelator concentration of 0.05 M, EDTAsolubilized more Fe than did PDTA, with values of 227 mg kg−1and102 mg kg−1, respectively.

Some researchers have found that co-dissolution of soil Ca isan important factor that can result in a low degree of heavy metalcomplexation by EDTA [20,21]. Another study showed that Ca2+ isthe main competitive cation, because CaCO3 is strongly dissolvedin the EDTA leaching solution under acidic conditions [20,25]; thus,concentrations of Ca2+ in the leaching solution are very high com-pared to those of the target heavy metals. However, at pH 7.0, PDTAsolubilized 25% less Ca and extracted 54% more Cu from the con-taminated soil than did EDTA (Figs. 6 and 8). These results canbe partly attributed to the low metal selectivity of EDTA, caus-ing potential chelation of all of the exchangeable cations presentin the soil [18], and partly to the selective complexing ability ofPDTA for Cu. Thus, the minerals present in soil may not interferewith PDTA extraction of Cu, confirming the potential applicabil-ity of PDTA for remediation of heavy metal-contaminated soilcontaining high concentrations of mineral elements such as Caand Fe.

4. Conclusions

Two new EDTA derivatives, BTDA and PDTA, were synthesizedand their metal–ligand complexation equilibrium constants andselective capabilities in aqueous media were investigated, alongwith those of EDTA and CDTA. Titration results showed that PDTAhad the highest stability constants for Cu and Ni and the highestoverall selectivity for trace metals over major cations. Equilibriumbatch experiments were conducted to evaluate the efficacy of theEDTA derivatives at extracting Cu, Zn, Ni, Pb, Ca, and Fe from a con-taminated soil. At pH 7.0, PDTA extracted 1.5 times more Cu thandid EDTA, but only 75% as much Ca. Although CDTA was a strongchelator of heavy metals, its overall selectivity was lower and com-

parable to that of EDTA. BDTA was the least effective extractantbecause its stability constants with heavy metals were low. PDTAis potentially a practical washing agent for soils contaminated withtrace metals, especially Cu.

dous M

A

dSVtaLT

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[EDTA washing solution for saturated-soil remediation of trace metal contami-

T. Zhang et al. / Journal of Hazar

cknowledgements

The project was supported by National Natural Science Foun-ation (No. 41171374), National Funds for Distinguished Youngcientists of China (No. 41225004), Guangdong Province Higherocational Colleges & Schools Pearl River Scholar Funded Scheme,

he Ministry of Environmental Protection of China (No. 201109020)nd the Research Fund Program of Guangdong Provincial Keyaboratory of Environmental Pollution Control and Remediationechnology (No. 2011K0007).

eferences

[1] X.P. Li, C.C. Huang, Environment impact of heavy metals on urban soil in thevicinity of industrial area of Baoji city, P.R. China, Environ. Geol. 52 (2007)1631–1637.

[2] D. Lestan, C.L. Luo, X.D. Li, The use of chelating agents in the remediation ofmetal-contaminated soils: a review, Environ. Pollut. 153 (2008) 3–13.

[3] O. Barrutia, C. Garbisu, J. Hernandez-Allica, J.I. Garcia-Plazaola, J.M. Becerril,Differences in EDTA-assisted metal phytoextraction between metallicolous andnon-metallicolous accessions of Rumex acetosa L., Environ. Pollut. 158 (2010)1710–1715.

[4] S. Tandy, R. Schulin, B. Nowack, Uptake of metals during chelant-assisted phy-toextraction with EDDS related to the solubilized metal concentration, Environ.Sci. Technol. 40 (2006) 2753–2758.

[5] G. Dermont, M. Bergeron, G. Mercier, M. Richer-Lafleche, Soil washing for metalremoval: a review of physical/chemical technologies and field applications, J.Hazard. Mater. 152 (2008) 1–31.

[6] W.H. Zhang, D.C.W. Tsang, I.M.C. Lo, Removal of Pb by EDTA-washing in thepresence of hydrophobic organic contaminants or anionic surfactant, J. Hazard.Mater. 155 (2008) 433–439.

[7] T.C.M. Yip, D.C.W. Tsang, K.T.W. Ng, I.M.C. Lo, Kinetic interactions of EDDS withsoils. 1. Metal resorption and competition under EDDS deficiency, Environ. Sci.Technol. 43 (2009) 831–836.

[8] D.Y.S. Yan, T.C.M. Yip, M.M.T. Yui, D.C.W. Tsang, I.M.C. Lo, Influence of EDDS-to-metal molar ratio, solution pH, and soil-to-solution ratio on metal extractionunder EDDS deficiency, J. Hazard. Mater. 178 (2010) 890–894.

[9] D.C.W. Tsang, W.H. Zhang, I.M.C. Lo, Copper extraction effectiveness and soildissolution issues of EDTA-flushing of artificially contaminated soil, Chemo-sphere 68 (2007) 234–243.

10] W.H. Zhang, D.C.W. Tsang, I.M.C. Lo, Removal of MDF and Pb from contam-inated soils by EDTA- and SDS-enhanced washing, Chemosphere 66 (2007)2025–2034.

11] W.B. Xia, H. Gao, X.H. Wang, C.H. Zhou, Y.G. Liu, T. Fan, X. Wang, Applicationof EDTA decontamination on soils affected by mining activities and impactof treatment on the geochemical partition of metal contaminants, J. Hazard.Mater. 164 (2009) 936–940.

12] W.H. Zhang, L.Z. Tong, Y. Yuan, Z.Y. Liu, H. Huang, F.F. Tan, R.L. Qiu, Influence ofsoil washing with a chelators on subsequent chemical immobilization of heavymetals in a contaminated soil, J. Hazard. Mater. 178 (2010) 578–587.

13] R.W. Peters, Chelating extraction of heavy metals from contaminated soils, J.Hazard. Mater. 66 (1999) 151–210.

14] W.H. Zhang, I.M.C. Lo, EDTA-enhanced washing for remediation of Pb- and/orZn-contaminated soils, J. Environ. Eng. 132 (2006) 1282–1288.

15] R.S. Tejowulan, W.H. Hendershot, Removal of trace metals from contaminatedsoils using EDTA incorporating resin trapping techniques, Environ. Pollut. 103(1998) 135–142.

16] C.E. Martinez, H.L. Motto, Solubility of lead, zinc and copper added to mineralsoils, Environ. Pollut. 107 (2000) 153–158.

17] M. Udovic, D. Lestan, EDTA leaching of Cu contaminated soils using ozone/UVfor treatment and reuse of washing solution in a closed loop, Water Air SoilPollut. 181 (2007) 319–327.

18] M.A.M. Kedziorek, A.C.M. Bourg, Solubilization of lead and cadmium during thepercolation of EDTA through a soil polluted by smelting activities, J. Contam.Hydrol. 40 (2000) 381–392.

19] G.A. Brown, H.A. Elliott, Influence of electrolytes on EDTA extraction of Pb frompolluted soil, Water Air Soil Pollut. 62 (1992) 157–165.

20] N. Papassiopi, S. Tambouris, A. Kontopoulos, Removal of heavy metals fromcalcareous contaminated soils by EDTA leaching, Water Air Soil Pollut. 109(1999) 1–15.

[

aterials 262 (2013) 464– 471 471

21] P. Theodoratos, N. Papassiopi, T. Georgoudis, A. Kontopoulos, Selective removalof lead from calcareous polluted soils using the Ca-EDTA salt, Water Air SoilPollut. 122 (2000) 351–368.

22] A. Polettini, R. Pomi, E. Rolle, The effect of operating variables on chelant-assisted remediation of contaminated dredged sediment, Chemosphere 66(2007) 866–877.

23] L. Di Palma, R. Mecozzi, Heavy metals mobilization from harbor sediments usingEDTA and citric acid as chelating agents, J. Hazard. Mater. 147 (2007) 768–775.

24] H.M. Nabhan, J. Vanderdeelen, A. Cottenie, Chelate behavior in saline-alkalinesoil conditions, Plant and Soil 46 (1977) 603–618.

25] L. Di Palma, P. Ferrantelli, Copper leaching from a sandy soil: mechanism andparameters affecting EDTA extraction, J. Hazard. Mater. B122 (2005) 85–90.

26] D.C. Harris, Quantitative Chemical Analysis, 4th ed., Freeman, New York, 1995.27] M.A. Santos, S. Gama, L. Gano, E. Farkas, Bis(3-hydroxy-4-pyridinone)–EDTA

derivative as a potential therapeutic Al-chelating agent. Synthesis, solutionstudies and biological assays, J. Inorg. Biochem. 99 (2005) 1845–1852.

28] A.F. Danil de Namor, D.A. Pacheco Tanaka, Thermodynamics of protonation andcomplexation of EDTA derivatives and metal cations in water, J. Chem. Soc.,Faraday Trans. (1998) 94.

29] L.P. Van Reeuwijk, Procedures for Soil Analysis, 3rd ed., International Soil Ref-erence and Information Centre, Wageningen, The Netherlands, 1992 (variouspagination).

30] D.F. Ball, Loss-on-ignition as an estimate of organic matter and organic carbonin non-calcareous soils, J. Soil Sci. 15 (1964) 84–92.

31] G.W. Gee, J.W. Bauder, Particle-size analysis, in: A. Klute (Ed.), Methods of SoilAnalysis, Part 1: Physical and Mineralogical Methods, Soil Science Society ofAmerican, American Society of Agronomy, Madison, WI, USA, 1986.

32] J.B. Wang, X.H. Qian, A series of polyamide receptor based PET fluorescent sen-sor molecules: positively cooperative Hg2+ ion binding with high sensitivity,Org. Lett. 8 (2006) 3721–3724.

33] Z.W. Mao, G. Liehr, R.V. Eldik, Structural and mechanistic information onthe reaction of bicarbonate with Cu(II) and Zn(II) complexes of tris(2-aminoethyl)amine. Identification of intermediate and product species, J. Chem.Soc., Dalton Trans. 10 (2001) 1593–1600.

34] G. Gran, Determination of the equivalent point in potentiometric titrations,Acta Chem. Scand. 4 (1950) 559–577.

35] C.F.G.C. Geraldes, A.M. Urbano, M.C. Alpoim, A.D. Shrry, K.T. Kuan, R.Rajagopalan, F. Maton, R.N. Muller, Preparation, physic-chemical characteri-zation, and relaxometry studies of various gadolinium (III)–DTPA–bis (amide)derivatives as potential magnetic resonance contrast agents, Magn. Reson.Imaging 13 (1995) 401–420.

36] Y.H. Zhou, M. Zhao, Z.W. Mao, L.N. Ji, Ester hydrolysis by a cyclodextrin dimercatalyst with a metallophenanthroline linking group, Chem. Eur. J. 14 (2008)7193–7201.

37] M. Zhao, L. Zhang, H.Y. Chen, H.L. Wang, L.N. Ji, Z.W. Mao, Effect of hydrophobicinteraction cooperating with double Lewis acid activation in a zinc(II) phos-phodiesterase mimic, Chem. Commun. 46 (2010) 6497–6499.

38] M. Zhao, H.L. Wang, L. Zhang, C.Y. Zhao, L.N. Ji, Z.W. Mao, Unexpected phospho-diesterase activity at low pH of a dinuclear copper-�-cyclodextrin complex,Chem. Commun. 47 (2011) 7344–7346.

39] Z. Jaszberenyi, I. Banyai, E. Briicher, K. Robert, K. Hideg, T. Kalai, Equilibrium andNMR studies on GdIII, YIII , CuII and ZnII complexes of various DTPA–N,N–bis(amid) ligands. Kinetic stabilities of the gadolinium(III) complexes, DaltonTrans. (2006) 1082–1091.

40] J.D. Martin-Ramos, Copper(II) and Nickel(II) chelates with dihydrogentrans-1,2-diaminocyclohexane-N,N,N′ ,N′-tetraacetate(2−) ion (H2CDTA2−).Synthesis, XRD structure and properties of [Cu(H2CDTA)]·H2O and[Ni(H2CDTA)(H2O)]·4H2O, Polyhedron 15 (1996) 439–446.

41] M.X. Wang, X.M. Meng, M.Z. Zhu, Q.X. Guo, A novel selective fluorescentchemosensor for Cu(II), Chinese Chem. Lett. 18 (2007) 1403–1406.

42] R.E. Smith, A.E. Martell, Critical Stability Constants: 1, 4, 5, and 6, Plenum Press,New York, 1974, 1976, 1982, 1989.

43] Z.L. Zou, R.L. Qiu, W.H. Zhang, H.Y. Dong, Z.H. Zhao, T. Zhang, X.G. Wei, X.D. Cai,The study of operating variables in soil washing with EDTA, Environ. Pollut. 157(2009) 229–236.

44] B. Nowack, R. Schulin, B.H. Robinson, Critical assessment of chelant-enhancedmetal phytoextraction, Environ. Sci. Technol. 40 (2006) 5225–5232.

45] M.D. Andrade, S.O. Prasher, W.H. Hendershot, Optimizing the molarity of a

nated soils, Environ. Pollut. 147 (2007) 781–790.46] B.R. Coughlin, A.T. Stone, Nonreversible adsorption of divalent metal ions (MnII,

CoII, NiII, CuII and PbII) onto goethite: effects of acidification, FeII addition, andpicolinic-acid addition, Environ. Sci. Technol. 29 (1995) 2445–2455.