Immobilization of heavy metals in polluted soils by the addition of zeolitic material synthesized from coal ﬂy ash

7/25/2019 Immobilization of heavy metals in polluted soils by the addition of zeolitic material synthesized from coal y ash

http://slidepdf.com/reader/full/immobilization-of-heavy-metals-in-polluted-soils-by-the-addition-of-zeolitic 1/10

Immobilization of heavy metals in polluted soilsby the addition of zeolitic material synthesized

from coal fly ash

Xavier Querol a,*, Andres Alastuey a, Natalia Moreno a,Esther Alvarez-Ayuso b, Antonio Garcıa-Sanchez b, Jordi Cama a,

Carles Ayora a, Mariano Simon c

a Institute of Earth Sciences ‘‘Jaume Almera’’ (CSIC) c/Lluis Sole i Sabarı s, s/n, 08028, Barcelona, Spainb Instituto de Recursos Naturales y Agrobiologı a (CSIC) c/Cordel de Merinas, 40, Apdo. 257, Salamanca 37071, Spain

c Departamento de Edafologı a y Quı mica Agrı cola de la Universidad de Almeria, Spain

Received 22 December 2004; received in revised form 12 May 2005; accepted 13 May 2005Available online 21 July 2005

Abstract

The use of zeolitic material synthesized from coal fly ash for the immobilization of pollutants in contaminated soilswas investigated in experimental plots in the Guadiamar Valley (SW Spain). This area was affected by a pyrite slurry

spill in April 1998. Although reclamation activities were completed in a few months, residual pyrite slurry mixed withsoil accounted for relatively high leachable levels of trace elements such as Zn, Pb, As, Cu, Sb, Co, Tl and Cd. Phyto-remediation strategies were adopted for the final recovery of the polluted soils. The immobilization of metals had pre-viously been undertaken to avoid leaching processes and the consequent groundwater pollution. To this end, 1100 kg of high NaP1 (Na6[(AlO2)6(SiO2)10] Æ 15H2O) zeolitic material was synthesized using fly ash from the Teruel power plant(NE Spain), in a 10 m3 reactor. This zeolitic material was manually applied using different doses (10000–25000 kg perhectare), into the 25 cm topsoil. Another plot (control) was maintained without zeolite. Sampling was carried out 1 and2 years after the zeolite addition. The results show that the zeolitic material considerably decreases the leaching of Cd,Co, Cu, Ni, and Zn. The sorption of metals in soil clay minerals (illite) proved to be the main cause contributing to theimmobilization of these pollutants. This sorption could be a consequence of the rise in pH from 3.3 to 7.6 owing to thealkalinity of the zeolitic material added (caused by traces of free lime in the fly ash, or residual NaOH from synthesis). 2005 Elsevier Ltd. All rights reserved.

Keywords: Coal fly ash; Zeolite; Immobilization; Metals; Polluted soils; Soil remediation; Aznalcollar mine spill

1. Introduction

High concentrations of heavy metals in soils maycause long-term risks to ecosystems and humans.Although heavy metals are released in varying quanti-ties into the soil from parent materials, increasingenvironmental contamination has been caused by

0045-6535/$ - see front matter 2005 Elsevier Ltd. All rights reserved.

doi:10.1016/j.chemosphere.2005.05.029

* Corresponding author. Tel.: +34 934095410; fax: +34934110012.

E-mail address: [email protected] (X. Querol).

Chemosphere 62 (2006) 171–180

www.elsevier.com/locate/chemosphere

mailto:[email protected]

mailto:[email protected]



human activities, such as mining, smelting, fossil fuelcombustion, agricultural practices and waste disposal(Ross, 1994; Alloway, 1995).

A pyrite slurry spill occurred in the Guadiamar river(a tributary river of the Guadalquivir river, SW Spain)on 25 April 1998 as the result of the collapse of a dam

in a pond containing pyrite slurry and waste water witha high content of potentially toxic elements (Grimaltet al., 1999). This pond is located at Aznalcollar(Sevilla), where the sulphide ore deposits were extracted.An area 40 km long and 0.5 km wide (following theAgrio and Guadiamar valleys) was covered by a 30 cmto 3 m thick layer of pyrite slurry.

After soil reclamation, persistent soil pollution waslocally detected due to the presence of a small fractionof pyrite mud, with high concentrations of trace ele-ments such as Cd, Co, Cu, Ni, Zn and As, mixed withsoil. Despite the natural capacity of soils to reduce solu-

bility and bioavailability of metals (Mitchell, 1993; Mar-osits et al., 2000; Miranda-Trevino and Coles, 2003) atmany of the most seriously contaminated sites, environ-mental risks persist, demanding immediate action.Chemical immobilization by means of soil amendmentshas been recently investigated as a valuable alternativetechnique for a wide range of contaminated sites (Vang-ronsveld and Cunningham, 1998). Chemical immobiliza-tion prevents the transport of contaminants into deepersoil layers and into groundwater. Moreover, revegeta-tion of highly polluted sites might be possible afterimmobilization of phytotoxic trace elements (Vangrons-veld et al., 1996).

A number of natural or synthetic materials, such ascarbonates, phosphate rocks, alkaline agents, zeolites,clay minerals and organic materials, have been recentlytested in order to evaluate their ability to immobilizetoxic trace metals. These amendments can lead to theimmobilization of metals in a variety of ways. Firstly,some amendments dissolve supplying alkalinity to theacid polluted soil, causing the precipitation of insolublephases. These neoformed phases contain metals as majorconstituents, such as metal-phosphates and carbonates(Chen et al., 2000; Basta et al., 2001; Seaman et al.,2001), or as minor components co-precipitated in

hydroxides (Chlopecka and Adriano, 1996; Vangrons-veld and Cunningham, 1998; Boisson et al., 1999).

Secondly, the increase in alkalinity promotes themetal sorption via surface complexation processes. Min-eral surfaces have a positive charge at low pH values dueto the sorption of protons, and they acquire a negativecharge as pH increases owing to the deprotonation of the surface unsaturated bonds. Therefore, as pH in-creases cations form stable complexes with the negativeradicals on the surfaces. Among soil minerals, clays playa significant role in surface complexation because of their higher specific surface (Du et al., 1997a,b). How-

ever, since As forms anionic species in solution, amend-

ments increasing the soil pH may also result in Asmobilization (Hingston et al., 1971; Golberg, 1986; Gol-berg and Glaubig, 1988).

Thirdly, metal retention may also take place regard-less of pH due to the cation exchange in zeolites (Nissenet al., 2000; Castaldi et al., 2005). Zeolites are crystalline

aluminum-silicates, with group I or II elements as coun-ter ions. Their structure is made up of a framework of [SiO4]4 and [AlO4]5 tetrahedra linked to each otherat the corners by sharing their oxygens. The substitutionof Si(IV) by Al(III) in the tetrahedra accounts for a neg-ative charge of the structure which may give rise to ahigh cation exchange capacity (CEC) (up to 5 meq/g)when the open spaces allow the access of cations (Breck,1984). Zeolites may be found in natural deposits, gener-ally associated with the alkaline activation of glassy vol-canic rocks, or synthesized from a wide variety of high Siand Al starting materials.

Since the initial studies by Holler and Wirsching(1985), a number of patents and technical articles haveproposed different hydrothermal activation methods tosynthesize diverse zeolites from coal fly ash. All themethodologies developed are based on the dissolutionof Al–Si-bearing fly ash phases with alkaline solutions(mainly NaOH and KOH solutions) and the subsequentprecipitation of zeolitic material. Using this procedure,zeolites A and X, KM (equivalent to phillipsite), NaP1(Na6[(AlO2)6(SiO2)10] Æ 15H2O), Na-chabazite (herschel-ite), K-chabazite, Linde F, and other high CEC zeoliteshave been obtained from coal fly ash.

This study focuses on the application of NaP1 zeo-litic material, synthesized from coal fly ash from the Ter-uel power plant (ENDESA, NE Spain), to the pollutedsoils to diminish the leachable contents of heavy metalsduring phyto-remediation. The alkalinity of the zeoliticproduct (due to residual NaOH from synthesis, or tothe occurrence of free lime in relict fly ash particles)may also contribute to buffer the acidity of the soil,and consequently may also favour plant growth.

2. Materials and methods

2.1. Synthesis of zeolitic products

Zeolitic products were obtained at pilot plant scale(at CLARIANT SA) by alkaline conversion of coal flyash from the Teruel power plant (NE Spain, ENDESA)(Querol et al., 2001). The Teruel fly ash was selected toreproduce the laboratory tests at pilot plant scale byusing a low-glass fly ash (see major characteristics of thisfly ash in Table 1). The synthesis conditions were se-lected from prior studies (Querol et al., 1997) to obtainhigh NaP1 products (Na6[(AlO2)6(SiO2)10] Æ 15H2O)owing to the high CEC values of the NaP1 zeolite

(5 meq/g). The experiments were carried out in a 10 m

3

172 X. Querol et al. / Chemosphere 62 (2006) 171–180



R-410-A reactor made of 304 stainless steel. The experi-mental parameters used for the synthesis of 1100 kg of this zeolitic material in a single batch were: 2 l of 2 MNaOH/kg fly ash during 24 h, at 150 C (Querol et al.,2001). The zeolitic material synthesized was filtered (55pneumatic press filters, 90 C during 1 h) and washedwith water (at 85 C during 40 min), dried at room tem-perature and analyzed by means of X-ray diffraction

(XRD) with CuKa radiation. A semi-quantitative esti-

mation of the zeolite contents was obtained comparingtheir CECs with that of a pure NaP1 commercial product(produced by Industrias Quimicas del Ebro, IQE). TheCEC of the zeolitic materials was calculated using ammo-nium solutions following the methodology of the Interna-tional Soil Reference and Information Centre (ISRIC,

1995). A CEC value of 2.0 meq/g obtained for the finalproduct points towards a zeolite content of around 45%(when compared with the 5 meq/g obtained for the IQEcommercial zeolite). In addition to the zeolite, the finalproduct contains relict fly ash particles made up of quartz, mullite, Al and Si glass, magnetite, traces of lime,and residual NaOH from synthesis.

2.2. Soil properties

Representative samples of the contaminated soil werecollected to determine the main soil properties (Table 2).

The <2 mm fraction was used with this aim. The miner-als present in the sample were determined by XRD pat-terns and the proportions quantified by the Rietveldmethod (Rodrıguez-Carvajal et al., 1987). The samplewas made up of quartz, illite, montmorillonite, clinoch-lore, actinolite, microcline + albite and gypsum. The pHwas determined potentiometrically in a soil paste satu-rated with water. Organic matter was determined bydichromate oxidation using the Tiurin method (Jackson,1960) and particle size distribution was analysed by thepipette method (Gee and Bauder, 1986). Contents of Cand N were determined in the control soil with an ele-mental analyzer. Soil sub-samples of the <260 lmfraction were used for the total content metal determina-tion. Accurately weighed amounts of each sample wereacid digested following a two-step dissolution procedure

Table 1Major chemical and physical characteristics of the Teruel flyash

Major oxides (%)

SiO2 48.3Al2O3 23.9

Fe2O3 16.0CaO 5.4MgO 1.0Na2O 0.2K2O 1.4P2O5 0.2TiO2 0.8SO3 0.8SiO2/Al2O3 2.0

Trace elements (l g/g)

As 79B 342Ba 311

Cr 107Cu 52Li 256Mo 15Pb 65Sr 523U 20V 206Zn 174

Mineral composition (%)

Glass 62.7Mullite 19.4Quartz 8.6

Anhydrite 1.5Lime 0.7Hematite 5.9Magnetite 1.3Feldspar <1.0

Physical characterisation

Grain size (lm)P10% 4.6Median 21.8P90% 75.7

True density (g/cm3) 2.5Apparent density (g/cm3) 1.1Porosity (%) 58.2

BET SA (m2

/g) 1.9P10% and P90%: percentile 10% and 90%, respectively.

Table 2Selected soil properties

Properties Soil control

pH 3.5–3.9Organic matter (%) 0.80Carbon (%) 0.46Nitrogen (%) 0.060

Mineralogy (% from XRD analysis)

Quartz 30Illite 20Clinochlore 5Montmorillonite 5Actinolite 5Feldspars 25Gypsum 8

Grain size distribution (%)

250–2000 lm 45.550–50 lm 31.050–2 lm 12.7<2 lm 8.8

X. Querol et al. / Chemosphere 62 (2006) 171–180 173



(a HNO3 extraction followed by a HNO3:HF:HClO4

extraction) (Querol et al., 1995). The contents of majorand trace elements in the solutions obtained from thetotal sample digestion were analysed by (a) inductivelycoupled plasma atomic emission spectrometry (ICP-AES) for major and selected trace elements (Ca, Al, S,

Fe, Mg, Na, Mn, Ba, Cu, P, Ni, and Zn) and (b) induc-tively coupled plasma mass spectrometry (ICP-MS) formost of the trace elements (As, Ba, Cd, Co, Cu, Cr,Mo, Ni, Pb, Sr, Th, Ti, Tl, V and Zn). Analytical accu-racy was checked with SARM 19 and 1633b referencematerials and yielded analytical errors <3% with theexception of P and K (with 10%).

2.3. Soil treatment with zeolite

In order to test the immobilization potential of thezeolite material for the leachable trace elements, on

29th April 1999, the zeolitic product was dosed inproportions of 10000, 15000 and 25000 kg/ha andmanually mixed with the 25 cm top soil in three experi-mental plots (20 m2 each) around the Vicario farm(Sanlucar la Mayor village). A control plot was keptas a reference.

The top 15 cm of soil from each field were sampledby collecting two continuous 20 cm wide channel soilsamples connecting the corners of the square fields diag-onally. The soil material collected was mixed and riffledto obtain a final sample of 10 kg which was transportedto the laboratory for analysis. Sampling of treated andcontrol soils was carried out in 27th April 2000 and23rd March 2001. Five replicates were sampled fromeach field to tests the representativity of the sampling.Relative standard deviations (RSD) of different analyseswere lower than 40% for most elements as later will beshown.

2.4. Leaching studies

2.4.1. Batch tests

Accurately weighed amounts of each soil sample of the <260 lm fraction were leached following the DIN38414-S4 (1984) procedure with the aim of determining

the levels of water extractable elements (100 g of samplein 1 l MilliQ grade water, stirring under room tempera-ture, 24 h in a PVC container). The contents of majorand trace elements in the leachates were analysed byICP-AES and ICP-MS. Furthermore, the conductivityand pH were measured in the leachates using conven-tional procedures.

2.4.2. Column tests

In addition to the above closed leaching procedures,the mobility of metals as a function of time was studiedas water was added, in an open leaching test, using glass

columns packed with 100 g of either untreated soil (con-

trol) or soil amended with zeolite (15000 kg/ha,25000 kg/ha). The columns (3 · 15 cm) were leachedwith 400 ml (560 mm annual rainfall in this area) of de-ionized water under a saturated flow regime, collect-ing successive leached fractions of 25 ml using a fractioncollector (Foxy Jr., ISCO). Metal concentrations in the

leached fractions were determined by atomic absorptionspectrometry (AAS).

In order to confirm possible ion exchange processes,the readily extractable metal fraction was determined inthe soil, before and after treatment. The DIN V 19730(1993) procedure was used to this end. In this methoda soil mixture with a soil: extractant ratio of 1:2.5 is stir-red for 2 h using 1 M NH4NO3 as extracting agent.Metal concentrations in the extracts were analyzed byAAS.

2.5. Effect of pH variation

To assess the variation of the soil retention capacitywith pH regardless of zeolite exchange processes, a seriesof batch experiments were conducted at room tempera-ture (21 ± 2 C) to examine the variation of metal aque-ous concentration (Fe, Al, Cu, Zn, Cd, Co, Ni and Mn)as a function of pH. Experiments consisted of mixing 1 gof soil sample (grain fraction between 30 and 100 mm)into 100 ml of desired pH solution, ranging from 3.5to 7.5. Solution pH in the range of 3.5–5.5 was madeup of HCl solutions, while pH higher than 5.5 was main-tained constant by adding different amounts of 1 MNa2CO3. The mixture was stirred for 2 h to reach equi-librium. Afterwards, 25 ml of suspension were filteredthrough a 0.45 mm Millipore filter and ICP-MS wasused to analyse the contents of trace elements.

3. Results and discussion

3.1. Bulk and leachable metal concentrations

Table 3 gives the results of the bulk content of majorand trace elements in the original soil, and the conduc-tivity, pH and leachable concentrations (DIN 38414-

S4) of major and trace elements of the samples obtainedin the 27/04/2000 and 23/03/2001 sampling periods (1and 2 years after the addition of NaP1-TE). Accordingto the Dutch legislation concerning soil and groundwa-ter contamination by toxic substances (NetherlandsMinistry of Housing, Physical Planning and Environ-ment, 1995) the total concentration of As surpassesgreatly the standard intervention value (55 mg/kg), indi-cating high As soil pollution and need of its remediation.Nevertheless, the percentage of As extractable withwater appears very low (<0.15%, 2 years after theflood), therefore the risk of leaching and groundwater

contamination with this element is really low in such






conditions. The total concentrations of Cd, Cu and Zn,although below the standard intervention values (12, 190and 720 mg/kg, respectively) are close to them, espe-cially in the case of Cu, suggesting that their controlshould be suitable as well. Although we recognize thatthe major pollution problem is As, we selected these pol-

luted soils to test the effect of the zeolite dosage on thereduction of levels of leachable metals.

Relative standard deviations (RSD) for bulk concen-trations of major and trace elements in the replicatessampled in the reference soil reached values of 3% forAl and Na; 8–15% for Ca, Fe, Mn, Mg, Ba, Co, Cu,Sr, V and Zn; 17–40% for Cr, Ni, P, S, As, Pb, Th;and 45–100% for Ti, Cd, Tl and Mo. RSDs of replicatesfor the leachable contents of trace elements in the refer-ence soil varied from 3% to 17% for most elements. Inthe amended soils these RSD values are only slightly in-creased (7–20% for most elements), with the exception of

Zn and V with relative standard deviations from 20% to65%.

Fig. 1b and c shows the percentage of water extract-able concentrations (% element leachable in water withrespect to the total content) for six elements dependingon the zeolitic inputs of the doses of the zeolitic product.The results demonstrated that 2 years after the flood,elements occurring in the pyrite slurry, such as Cd,Co, Cu, Ni, and Zn, were leached in a very high propor-tion. In the control soil the leachable contents of theseelements reached 45% Cd, 28% Co, 7% Cu and Ni,

and 53% Zn (Fig. 1). The mobility of these elementsdrastically decreased down to <1% after 1 and 2 yearsof the NaP1-TE zeolite addition with relatively lowdoses (15000 and 25000 kg/ha).

Concentrations of metals in leachates obtained withthe DIN procedure (solid/extractant ratios of 100 g/l)

applied to the control samples for the two sampling peri-ods reached 14692 ± 1758 and 16900 ± 2331 lg Zn/l,805 ± 69 and 1130 ± 798 lg Cu/l, 282 ± 40 and 280 ±36 lg Co/l, 4 ± 2 and 60 ± 4 lg Pb/l, 165 ± 22 and220 ± 27 lg Ni/l, 56 ± 4 and 80 ± 3 lg Cd/l; and 20 ±2 and 40 ± 14 lg As/l. In the field, with a zeolite dosageof 25000 kg/ha, the water extractable contents are re-duced down to 37 ± 13 and 10 ± 6 lg Zn/l, 28 ± 8 and20 ± 6 lg Cu/l, 9 ± 2 and 10 ± 2 lg Ni/l, 15 ± 2 and20 ± 2 lg As/l and <1 lg/l of Cd, Co and Pb.

As shown in Table 3 and Fig. 1, the control field hada pH of 3.5 and 3.9 after 1 and 2 years of the remedia-

tion activities, whereas the treated soils reached a pHof up to 7.5 and 8.0 as a consequence of the alkalinityof the zeolitic material. The buffering of the acidity of the soils may also result in the precipitation of some of the metals or in the adsorption of cations on clay sur-faces. Prior studies of heavy metal sorption from acidmine waters using NaP1 and 4A zeolites synthesizedfrom fly ashes have demonstrated that the combinationof the precipitation and ion exchange processes accountsfor the following affinity of the elements with respect tothe zeolite addition: Fe3+ = Al3+ > Cu2+ > Pb2+ > Cd2+ =

a

0

2

4

6

8

0 5000 10000 15000 20000 25000 30000

p H

0

500

1000

1500

2000

2500

( µ S / cm )

pH Conductivity

b c

April 2000

0

20

40

60

0 5000 10000 15000 20000 25000

Zeolite dose (kg/ha)


L e a c h a b l e ( % ) Cd

Co

Cu

Mn

Ni

Zn

March 2001

0

20

40

60

0 5000 10000 15000 20000 25000


Cd

Co

Cu

Mn

Ni

Zn

Fig. 1. pH and conductivity (27/04/2000) (a) and water extractable proportions for metals (% of element leachable in water with

respect to the total content) as a function of the zeolite dosing after 1 and 2 years (27/04/2000 (b) and 23/03/2001 (c), respectively).




Tl+ > Zn2+ > Mn2+ > Ca2+ = Sr2+ > Mg2+ (Morenoet al., 2001a,b). It should therefore be pointed out thatthe zeolitic product can selectively uptake metals in ahigh Ca and Mg media.

The experimental fields with the addition of the zeo-litic product showed significant plant growth, with re-

spect to the control field, probably as a consequence of the heavy metal sorption and the pH buffering effect of the zeolitic material. The following plant species wereidentified: Lupinus angustifolious, Oxalis pes-caprae,

Lamarckia aurea and Mentha piperita.

3.2. Soil percolation tests

The results of the open leaching tests (column tests)applied to the control soil and to the soils amended withzeolite (at doses of 15000 and 25000 kg/ha) are pre-sented in Fig. 2 for Cd, Co, Cu, Ni and Zn. In the case

of the control soil, percentages of the metal leached withrespect to the total metal content are 13% for Cd, 16%for Co, 9% for Cu, 4% for Ni and 19% for Zn. Most

of the eluted amounts of Cd, Co, Ni and Zn (>50%)were leached in the first two fractions (50 ml). Subse-quently there was a progressive decrease in the leachedamounts. As shown in Fig. 2, the percolation curvestend to reach the ‘‘plateau’’, indicating that the waterreachable fraction of the metals was almost completely

leached (for the current stage of the weathering of sul-phide phases). In the case of Cu, percolation curves(Fig. 2) suggest a less effective leaching, probably be-cause of its lower mobility due to its strong sorptionon clays when pH is increased (McBride, 1994).

In the case of the soil amended with the zeolite doseof 15000 kg/ha, the leached fractions of metals accountfor 1.1% of the total content of Cd, 0.6% of Co, 0.1% of Cu, 0.6% of Ni and 0.9% of Zn. For the highest zeolitedose (25000 kg/ha) these proportions are reduced to0.2% for Cd, Co, and Ni, and <0.03% for Co and Zn.Thus the amendment considerably reduced the leached

metal amount, to about 1–10%, even at the lowest zeolitedose applied to soil. The leaching rate decreased alongthe percolated water volume. Moreover, the cumulative

NiNi

Zn

Ni

Zn

NiNiNiNi

Zn

Soil (15 tonne/hectare)

L e a c h e d %

Cd

0

3

6

9

12

15

0 200 400 600

Co

0

3

6

9

12

15

18

0 200 400 600

Cu

0

2

4

6

8

10

0 200 400 600

Ni

0

1

2

3

4

5

0 200 400 600

Zn

0

5

10

15

20

25

0 200 400 600

Water volume (ml)

Soil (control)

Soil (25 tonne/hectare)

Water volume (ml)

L e a c h e d %

L e a c h e d %

L e a c h e d %

L e a c h e d %

Fig. 2. Cumulative curves of metal cations leaching in the soils.




curves took on a more concave form, reaching even the‘‘plateau’’ at the highest amendment dose indicating that,in this case, the leachable fraction of metal was com-pletely extracted.

In the light of these results it could be concluded thatzeolite amendment decreases the leaching of metals,

thereby eliminating or considerably minimizing the riskof groundwater pollution. On the other hand, the amountof As leached in the untreated soil (Fig. 3) accounts foronly 0.003% of the total As in the control soil. Althougha continuous leaching is deduced from the As percolationcurves, the low extraction values indicate a low mobilityfor As in the soil. The amounts of As leached are rela-tively higher for the soil samples amended with zeolite,obtaining 0.02% for the dose of 15000 kg/ha, and0.06% for 25000 kg/ha. Given the solid/extractant ratioof 100 g/400 ml used in the column leaching tests, thesevalues are equivalent to a concentration of <22 lg/l.

There is a marked correlation between the degree of As leaching and the rate of dosage of zeolitic material.One explanation for this could be the decrease in theAs sorption capacity of soil when pH is raised (up to7.4–8.0) by zeolite addition.

The major mechanism of As release from sulphideassemblages is by means of oxidation of arsenopyrite(Dove and Rimstidt, 1985). Subsequently, scorodite(FeAsO4 Æ 2H2O) may precipitate from an acidic, highlyconcentrated solution of Fe3+ and AsO3

4 (Chukhantsev,1956; Nishimura and Tozawa, 1978). Scorodite is meta-stable above pH 2 and tends to dissolve incongruently(Robins, 1987). The low pH (around 3–4) of soil samplesfavours As adsorption on Fe oxyhydroxides by means of ligand exchange; low pH causes the protonization of surface, transforming –OH to –OHþ

2 groups, which areeasier to displace from metal binding sites (McBride,1994). Several studies on the As adsorption on oxides,oxyhydroxides and soils provide evidence of largeadsorption capacities of oxyhydroxides with a maximum

adsorption around pH 3–5 (Golberg and Glaubig, 1988;Singh et al., 1996; Manning and Goldberg, 1997; Ravenet al., 1998; Smith et al., 1998; Garcia-Sanchez et al.,2002). The rise in the pH values due to the zeolite addi-tion may account for the reduction in the As sorptioncapacity of the soil.

An additional explanation for the increase of Asmobility due to zeolite addition is that the zeolite is asource of As. Thus, DIN-leaching tests performed forthe zeolitic material alone showed that 17 mg/kg of Ascould be released from the relict fly ash particles withpure water.

3.3. Metal de-sorption test

The heavy metal load still retained by the treated soilsamples (15000 kg/ha, 25000 kg/ha) after the NH4NO3

extraction for the sampling of 27/03/00 demonstrated

that the metal amounts extracted from zeolite are lowerthan 1% for all the metals studied except for Co (12%).

Both the low NH4-exchangeable metal amounts, aswell as the results of the batch experiments carried outwith alkaline solutions, suggest that, under the condi-tions and at the concentrations of the soil remediationtreatment, most of the retained metal load is not sorbedby the zeolitic material in an exchangeable form. There-fore, the depletion in the leached metal amounts ob-served as pH is increased must be due to an alternativeprocess. Metal adsorption onto illite surfaces and pre-cipitation of some hydroxides could account for themetal retention. To confirm this hypothesis, pH neutral-ization tests with soil samples were conducted in theabsence of zeolite. Fig. 4 shows the decrease in concen-tration of aqueous metals in soil solution (1 g/100 ml) asthe pH is increased from 3.5 to 7.5 in the soil sample. Toexplain this phenomenon, Cama et al. (2005) haveproposed a surface complexation model accounting formetal adsorption on illite since this phase is the predom-inant clay in the soil and has a high specific surface (Duet al., 1997a,b). Fig. 4a and b show that the experimentaldepletion of divalent metals matches the results obtainedfrom the surface complexation model. Precipitation of divalent metal solid phases such as oxides, hydroxides,

carbonates or sulfates was ruled out since these phaseswere always under-saturated in the solution.

With respect to trivalent metals (Fig. 4c), the de-crease in the Fe concentration could not be simulatedby using the sorption model, but could be simulatedby ferrihydrite precipitation. Decrease in Al concentra-tions could be modeled with the surface complexationon illite, although precipitation of gibbsite could notbe discarded at pH higher than 5.

It has been deduced that most metals are chemicallyimmobilized due to surface complexation (and hydroxideprecipitation for Fe and Al) as pH is increased. There-

fore, the zeolite addition has more effect on the immobi-

0.00

0.02

0.04

0.06

0.08

0.10

0 100 200 300 400 500

Water volume (ml)

A s l e a c h e d ( % )

control

15 tonne/hectare

25 tonne/hectare

Fig. 3. Cumulative curves of As leaching in the soils.




lization of these metals by the subsequent increase in pHthan by the zeolite cation exchange capacity. Thus, sim-ilar results in metal retention can be expected if otheralkaline amendments, such as lime, are used.

4. Conclusions

The zeolitic material synthesized from coal fly ash ap-

peared as an effective amendment to attenuate the soil

contamination by trace elements. After such a treatmentthe leachability of most metals (Cd, Co, Cu, Ni, Zn) de-creased around 95–99% when zeolite doses of 25 000 kg/ha were applied to soil. The soil pH increase caused bythe zeolite addition seems to be the factor responsiblefor metal immobilization, favoring metal adsorption

onto illite surfaces and precipitation of metal hydrox-ides. Although in case, the cation exchange process onthe zeolite is low, its use in the remediation of this pol-luted soil has two clear advantages with respect to otheralkaline agents: (a) the induced soil pH increase is mod-erate and therefore the risk of mobilization of elementssoluble in basic conditions is less than when using morealkaline agents and, (b) the zeolitic material, in additionto the buffering capacity, possesses an important ex-change capacity that could reduce the soluble metal con-centration if necessary; in this case, the amount of illitepresent in soil seemed enough to almost eliminate the

pollutant leachability.

Acknowledgments

The present study was supported by the BRITE-EURAM Program from the 4th Framework of R&Dof the European Union (SILEX, BRPR-CT98-0801)and by the Spanish CICYT (AMB99-1147-C02-02).We are indebted to the power generation company EN-DESA for supplying the fly ash sample from the Teruelpower plant, to Clarian SA for the pilot plant scale

experiment carried out to provide the zeolitic materialand to IQE SA for providing the pure commercial zeo-lite for comparison.

References

Alloway, B.J., 1995. Heavy Metals in Soils. Blackie Academic& Professional, Glasgow.

Basta, N.T., Gradwohl, R., Snethen, K.L., Schroder, J.L., 2001.Chemical immobilization of lead, zinc and cadmium insmelter-contaminated soils using biosolids and rock phos-phate. J. Environ. Qual. 30, 1222–1230.

Boisson, J., Mench, M., Vangronsveld, J., Ruttens, A., Kop-ponen, P., Koe, T., 1999. Immobilization of trace metalsand arsenic by different soil additives: evaluation by meansof chemical extractions. Commun. Soil Sci. Plant Anal. 30,365–387.

Breck, D.W., 1984. Ion exchange reactions in zeolites. Chapter7 of Zeolite Molecular Sieves, Structure, Chemistry, andUse. Robert E. Krieger Publishing Company, Malabar,Florida, TIC: 245213.

Cama, J., Ayora, C., Querol, X., Moreno, N., 2005. Metaladsorption on clays from a pyrite contaminated soil. J.Environ. Eng. 131, 1052–1056.

Castaldi, P., Santona, L., Melis, M., 2005. Heavy metal

immobilization by chemical amendments in a polluted soil

0E+00

1E-05

2E-05

3E-05

4E-05

[ c a t i o n s ] ( m o l / l )

Zn exptZn ads

Mn adsMn expt

Cu adsCu expt

0E+00

1E-07

2E-07

3E-07

4E-07

5E-07

6E-07

[ c a t

i o n s ] ( m o l / l )

Co expt

Co adsNi expt

Ni ads

Cd ads

Cd expt

0E+00

1E-05

2E-05

3E-05

4E-05

5E-05

6E-05

3 5 6 7 8

pH

[ A l

& F e ] ( m o l / l )

Al expt

Al prep

Al ads

Fe expt

(x10)

(x10)

Al (gibbsite)

Fe(ferrihydrite)

a

b

c 4

Fig. 4. Variation of aqueous metal concentrations as a functionof pH. The symbols are experimental (expt) values and thecurves represent the values calculated (ads) with a surfacecomplexation model (Cama et al., 2005). The Al and Feconcentrations in equilibrium with hydroxides (gibbsite andferridryte, respectively) are also plotted. Concentrations of Cuand Co have been enlarged by one order of magnitude forinclusion in the plot.




Documents

Immobilization of heavy metals in polluted soils by the addition of zeolitic material synthesized from coal ﬂy ash