Cement–clay pastes for stabilization/solidification of 2-chloroaniline

Cement–clay pastes for stabilization/solidification of 2-chloroaniline

Donatella Bottaa,*, Giovanni Dotellia, Riccardo Biancardia, Renato Pelosatob,Isabella Natali Sorab

aDipartimento di Chimica, Materiali e Ingegneria Chimica ‘‘Giulio Natta’’, Politecnico di Milano, Piazza L. da Vinci 32, 20133 Milano, ItalybINSTM and Laboratorio di Chimica Strutturistica, Dip. di Ingegneria Meccanica, Universita di Brescia, via Branze 38, 25123 Brescia, Italy

Accepted 21 October 2003

Abstract

Immobilization of a model liquid organic pollutant, i.e. the 2-chloroaniline (2-CA), into a cement matrix using organoclays aspre-sorbent agents was investigated. Five cement–clay pastes were prepared with different nominal water-to-cement ratios (w/

c=0.40, 0.25 and 0.15 wt/wt) and various amounts of waste (waste-to-cement o/c=0.20, 0.60 and 1.00 wt/wt); for comparison, aneat cement paste was also prepared. Dynamic leach tests were performed on solidified monoliths in order to assess the successfulimmobilization of the 2-CA. In monoliths at constant w/c ratio (0.40) the total amount of pollutant released increases with its initial

content, and ranges from 15 to 35% with respect to it. By lowering w/c from 0.40 to 0.15 at constant o/c, the performances improved(<25% released). The microstructure of the hardened cement–clay pastes was characterized by quantitative X-ray diffraction(QXRD) and electronic microscopy (SEM-EDS) techniques; hydration degree was estimated by means of thermogravimetric ana-

lysis (TGA) in addition to QXRD. No evidence of any chemical reaction between 2-CA and cement phases was found. Moreover, itwas shown that the most important factors affecting the cement hydration process were the total water content, i.e. the one takingalso into account the water contained in the wet polluted clay, and the amount of 2-CA not firmly sorbed by the organoclay, andthen freely dispersed in the paste.

# 2003 Elsevier Ltd. All rights reserved.

1. Introduction

Cement-based stabilization/solidification (S/S) pro-cesses have already proved their worth in the treatmentof heavy-metal containing hazardous wastes (Ouki andHills, 2002; Hills et al., 1999; Andac and Glasser, 1998;Diet et al., 1998; Hamilton and Bowers, 1997) and low-level radioactive wastes (Osmanliogu, 2002). For theimmobilization of organic liquid wastes in cementmatrices highly effective adsorbent materials are usedbefore the S/S treatment, in order to firmly bind theorganics to the cement matrix. If organics werestraightforwardly admixed with cement, they wouldaffect the cement hydration kinetics by retarding thereactions via formation of a protective film around thecement grain, hindering the formation of calcium

hydroxide, and accelerating the reaction via modifi-cation of the colloidal C–S–H (C¼CaO, S¼SiO2,H¼H2O) gel precipitated at very early stages around thecement grains (Chandra and Flodin, 1987; Pollard et al.,1991; Abd El Wahed, 1991; Edmeades and Hewlett,1998). For example, as we have reported in a previouswork (Natali Sora et al., 2002) for 2-chloroaniline (2-CA),a compound selected as representative of chloroaromaticamines, a mixture of methanol and 2-CA retards settingand hardening of cement pastes, however the usualreactions products do not seem to be altered. Moreover,2-CA is not firmly immobilized in the cement matrix.Organically-modified clays (organoclays) obtained by

exchanging naturally occurring cations (Na+, K+,Ca2+, Mg2+, etc.) with organic cations, usually fromquaternary ammonium salts (QAS) bearing long alkylchains, are able to adsorb organic chemicals, likephenols, chlorinated phenols, benzene, toluene, ethyl-benzene, xylene, and others (Montgomery et al.,1991; Sheng et al., 1996; Lo et al., 1997; Jaynes andVance, 1999) from water. The sorption properties oforganoclays have suggested their use in hazardous

0956-053X/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.

doi:10.1016/j.wasman.2003.10.005

Waste Management 24 (2004) 207–216

www.elsevier.com/locate/wasman

* Corresponding author. Tel.: +39-02-2399-3234; fax: +39-02-

70638173.

E-mail addresses: [email protected] (D. Botta), giovanni.

[email protected] (G. Dotelli), [email protected]

(R. Pelosato), [email protected] (I. Natali Sora).

http://www.elsevier.com/locate/wasman/a4.3d

mailto:[email protected]





waste remediation as geosorbents, for example, indecontaminating water and in retaining the adsorbedorganic contaminants for a long time (Lo, 1996). Themechanism controlling the sorption process is generallyclassified as ‘‘adsorption’’, where the solute is physicallyentrapped on the surface of the clay (Lee et al., 1990;Smith et al., 1990). When the solute is distributedbetween two phases, generally water and the organicphase in the interlayer space of the organoclay, themechanism is termed ‘‘partitioning’’ (Boyd et al., 1988;Smith et al., 1990; Dotelli et al., 2001).Azo-dyes, nitro-, and chloronitro-aromatic com-

pounds that, together with aromatic amines, are foundas pollutants resulting from improper disposal of textileindustrial wastes, in soil are ultimately transformed inaromatic amines (Brown and Laboreur, 1983), classifiedas hazardous materials. When dispersed in the environ-ment, the aromatic amines are generally sorbed in thesolid phase, where they can stay unchanged for a longtime. Because it is rather unknown the ultimate fateof aromatic amines and how long they could holdtheir harmful potential ought to their resistance to thebiological attack (Pagga and Brown, 1986), it is impor-tant to find a treatment for their detoxification and theelimination of the risk related to their environmentalpersistence.The aim of this work is to investigate a viable way to

immobilize, or anyway to detoxify, a class of hazardousorganic pollutants, i.e. aromatic amines, in a cement-based matrix. A model organic contaminant, 2-CA, pre-sorbed on organoclay, was mixed with an ordinaryPortland cement (OPC). X-ray diffraction (XRD) andelectron microscopy (SEM-EDS) techniques were usedfor the characterization of hardened cement pastes;thermogravimetric analysis (TGA) and QXRD to esti-mate the degree of cement hydration; dynamic leachtests were carried out in order to quantify the amount ofimmobilized organic compound.

2. Experimental

2.1. Sample preparation

Commercially available organoclay (Laviosa Chi-mica Mineraria s.r.l., Italy) was used to sorb theorganic compound. Organoclays (2.5 g) were weighedinto a centrifuge vial. 2-CA was added to obtainan aqueous suspension of 22 470 ppm in 50 ml ofMilliQWaters water. The vial was left on a rotatingarm for 24 h. Then, after centrifugation, 5 ml of theaqueous phase were extracted with 1 ml of iso-octane(RS Carlo Erba for spectroscopy). The extract wasanalysed by gas chromatography and the sorbed 2-CA was determined by difference: 337.9 g of 2-CA perkg of organoclay.

Still wet, the organoclay containing 2-CA was used,as model waste, for the preparation of all cement pastes.The 2-CA organoclay mixture is referred to as pollutedorganoclay, and the final monoliths will be addressed ascement–clay pastes. The initial concentration of 2-CAwas chosen taking into account the sorption capacity ofthe used clay: such a value belongs to the linear part ofthe sorption isotherm of the considered organoclay(Dotelli et al., 2001). The approximate water content ofthe polluted organoclay was estimated by mass balanceto be about 55 wt.%.OPC Type I 42,5 (UNI ENV 197) was used to prepare

samples; detailed raw oxide composition and Boguecalculation are reported in Natali Sora et al. (2002).Two groups of cement–clay pastes were prepared:

1. three pastes with a fixed water-to-cement ratio
(w/c=0.40) by varying polluted organoclay-to-cement ratio: o/c=0.20–0.60–1.00, samples A, B,and C, respectively;
2. two pastes with a fixed o/c ratio equal to 1.00 by
varying w/c ratio: 0.25 and 0.15; samples C0 andC00 respectively.
Moreover, a cement paste without clay, w/cratio=0.4, was prepared as reference (R). The compo-sition of the polluted organoclay is unvaried, i.e. 337.9 gof 2-CA per kg of organoclay, but the amount admixedwith cement powder changes from 12.5 (A) to 46.5 (C00)g per 100 g of paste; details are found in Table 1.Contaminated organoclay and cement binder were

hand-mixed with deionized MilliQWaters1 water, andthe fresh pastes were cast into PMMA moulds, partiallyfilled and sealed with Parafilm1. The pastes were curedin an air-conditioned room at 23�1 �C and 50% RH

Table 1

In the former group of columns the nominal compositions of cement–

clay pastes are reported

Sample P
astes composition P olluted organoclays (g/100 g of paste)
w
/ca o /cb a /cc O rganoclay W ater 2 -CA
R 0
.40 0 0 0 0 0
A 0
.40 0 .20 0 .022 4.2 6.9 1 .4
B 0
.40 0 .60 0 .067 1 0.0 1 6.6 3 .4
C 0
.40 1 .00 0 .112 1 3.9 2 3.1 4 .7
C0 0
.25 1 .00 0 .112 1 4.8 2 4.6 5 .0
C00 0
.15 1 .00 0 .112 1 5.5 2 5.8 5 .2
w indicates the water mixed with OPC-polluted organoclay mixtures, o

indicates the polluted organoclay mixed with OPC, and a indicates the

amount of 2-CA pre-sorbed on the organoclay. In the latter group, the

amounts of dry organoclay, water, and 2-CA of the model waste

admixed with cement powder are given per 100 g of paste initial

weight.a Nominal water-to-cement weight ratios.b Nominal polluted organoclay-to-cement weight ratios.c 2-CA-to-cement weight ratios.

208 D. Botta et al. /Waste Management 24 (2004) 207–216

for 28 days. Afterwards, at different curing times, i.e. 35and 70 days, one sample for each group was demoulded,dried up to constant weight at 85 �C for 8 h to stophydration. The weight loss after desiccation was mea-sured in order to estimate the free water content andreported as grams per gram of cement. Then the mono-lith was crushed, a fraction ground in an agate mortar;powders and fragments were stored in a desiccator forthe XRD, SEM-EDS, and TGA measurements.

2.2. Leach tests

In the present work the dynamic leach test (DLT) wascarried out in compliance with the UNI 8798 standard(seeUNICEN8798 in the references). The fresh pastes werecast into cylindrical PMMAmoulds (3.2 cm height and 2.2cm diameter), sealed and cured for 28 days in a room at23�1 �C and 50% RH. The leaching test was started after28-day curing: solidified samples were smoothed, cleanedand each of them hung in a 500-ml jar filled with 250 mlof deionized water, completely immersed and maintainedat 24 �C without agitation. Water was periodicallyrenewed according to the procedure: once per day thefirst week, twice the second week, once per week up to thesixth week, then once per month. These operations wereperformed for a number of progressive extractions up to110 days for the five series of samples (A, B, C, C0, andC00). Concentration of 2-CA in leachates was determinedat each leachant renewal, according to the UNI standardschedule; the amount of 2-CA was gas-chromato-graphically measured after isooctane extraction. GCanalyses were performed with a Carlo Erba Mega mod.5100 instrument, equipped with an on-column injector,flame ionization detector and HP fused silica capillarycolumn 0.32 mm internal diameter and 50 m length,coated with 5% phenylmethylsilicone rubber of 0.5 mmthickness. The temperature was linearly raised from 70to 130 �C at 4 �C/min; then to 250 �C at 10 �C/min. Afinal isothermal time of 5 min at 250 �C was maintained.

2.3. X-ray diffraction

The X-ray diffraction data were collected with a Phi-lips PW 1710 diffractometer under ambient conditionsusing graphite monochromated Cu-Ka radiation. Thestep scan was 0.02� and the measuring time 10 s perstep. The solidified specimens were ground in an agatemortar. The X-ray diffraction line profile analysis ofoverlapping peaks was made with WINFIT computerprogram (Krumm, 1997). The hydration of the cementpastes was estimated by calculating the amount of cal-cium hydroxide, which is a crystalline product of thehydration reactions of tricalcium silicate (Ca3SiO5) andb-dicalcium silicate (b-Ca2SiO4). The quantitative X-raydiffraction analysis was applied by using the equationproposed by Copeland and Bragg (1958) relating the

ratio of the integrated intensities (I0 and I1) of the X-raydiffraction lines of two components of a mixture tothe ratio of the weight fractions (w0 and w1) of thosecomponents

I1=I0 ¼ aw1=w0 ð1Þ

where a is a proportionality constant.

2.4. Electron microscopy

A scanning electron microscope (SEM) CambridgeStereoscan 260 equipped with a Link energy dispersivespectrometer (EDS) was used for the morphological andchemical analysis of the hydrated samples. The polishedsamples were gold coated to prevent charging effects.

2.5. Thermogravimetric analysis

Sample powders, previously dried and sieved (80-mesh), were heated at a rate of 10 �C/min up to 1000 �C(TG/DTA 6300 SEIKO). The amount of bound waterwas determined by the weight loss between 105 and1000 �C (loss on ignition, LOI). In this case, the pre-sence of wet clay containing a large amount of wastesuggested to express the results in grams per gram ofcement. However, special care is needed, because datahave to be corrected not only for the thermogravimetricloss of unreacted cement powder, but also for the lossesdue to 2-CA and organic part of the clay.

3. Results and discussion

3.1. Leaching behaviour

The inability of a cement matrix to retain organicsubstances is well known (Wiles, 1987). It was recentlydemonstrated that just the 2-CA is almost completelyreleased from solidified wasteforms (Natali Sora et al.,2002), when no pre-sorbent agent is used. On the con-trary, present DLT results show that the total amountof 2-CA released is never greater than about one thirdof the total amount originally contained in the waste-form (Fig. 1). Then, the addition of the organoclay as asorbent greatly improves the immobilization of 2-CAwith respect to the direct inclusion in the cement paste,where, in the most favourable case, more than 50 wt%was released, with peaks up to 90 wt.% (Natali Sora etal., 2002). As expected, at constant w/c ratio (samplesA, B, and C, Table 1) the larger the amount of 2-CA thegreater the quantity released (Fig. 1a). On the otherhand, at constant o/c ratio (sample C, C0, and C00,Table 1) amounts of 2-CA released from samples C0 andC00 are comparable and less than those leached from C.This suggests that the w/c ratio is an important factoraffecting the process effectiveness.

D. Botta et al. /Waste Management 24 (2004) 207–216 209

In order to better understand which factors influencethe immobilization, in Fig. 2a and b the initial 2-CAcontent and final cumulative amount released are com-pared. In Fig. 2a the results are reported as grams of 2-CA per gram of monolith initial weight. Sample C con-taining the largest amount of water, but an intermediatequantity of 2-CA (39.8 g of water and 4.7 g of 2-CA per100 g of monolith initial weight) has the maximumrelease (33%), while A, which contains the minimumamount of both (31.9 g of water and 1.4 g of 2-CA per

100 g of monolith initial weight), has the minimumrelease (13%). In Fig. 2b the results are reported asgrams of 2-CA per gram of cement: the less the amountof the binder the larger the amount of 2-CA released(sample C) and viceversa (sample A). In conclusion,binder, water and 2-CA contents are interrelated factorsand careful balancing between them is necessary toobtain the minimum release. However, a compromisehas to be reached between final weight of solidifiedwaste and content of sequestered pollutant.

Fig. 1. Cumulative amount of 2-CA released vs. leaching time: (a) (upper) reports samples A, B, and C (w/c=0.40), while (b) (lower) reports sam-

ples C, C0, and C00 (o/c=1.00); the reference sample R1 (w/c=0.40) containing 2-CA (1.2 g per 100 grams of initial weight) without organoclay is

also added (Natali Sora et al., 2002) for comparison. Data are referred to the total amount of 2-CA originally contained in the sample.


The workability of all pastes was good although the w/cratio in samples C0 and C00 was apparently extremely low;in fact, as deduced from Table 1, the total liquid/solidratio is quite large. This explains the good workability ofall pastes. Moreover, the porosity of a paste dependsstrongly on the liquid content. Since the samples weresealed during their curing, one may suppose that the largerthe liquid the higher the porosity. High porosity adverselyaffects the leaching behaviour of the solidified wasteforms,especially when the pollutant is not firmly immobilized.

Consequently, samples A and C gave the best and worstresults, lowest and highest porosity respectively.Although some organic admixtures simply modify the

rheological behaviour of the fresh binder mixture, mostinterfere with the hydration of Portland cement in dif-ferent ways (Wilding et al., 1984; Chandra and Flodin,1987; Abd El Wahed, 1991; Edmeades and Hewlett,1998). For these reasons, the assessment of a S/S pro-cess based on cementitious binders also requires thatmineralogical and microstructural features of solidified

Fig. 2. Total amount of 2-CA released during the whole leaching period, i.e. 110 days, compared with the initial content. Data are expressed in

grams of 2-CA per gram of monolith initial weight, i.e. cement–clay paste, (a) (upper), and in grams of 2-CA per gram of binder, i.e. Portland

cement, (b) (lower).


wasteforms be analysed. In the present work, the phasecomposition and the microstructure of the monolithswere investigated by QXRD, SEM-EDS and TGAtechniques at two different times: 1 week after thestarting of the leaching tests (35 days) and more than 1month later (70 days).

3.2. Mineralogy

The powder diffraction patterns from 70-day-oldhydrated pastes are shown in Fig. 3. In the XRD pat-terns the main clinker phases (alite, belite), calcite, and afew products of hydration reactions (calcium hydroxide,ettringite and/or thaumasite, Ca4Al2O6CO3

.11H2O andamorphous calcium silicate hydrate, C–S–H) were iden-tified. In our study no evidence of products from che-mical reactions involving 2-CA was found. Thehydration of the pastes was estimated by using Eq. (1).To determine the amount of calcium hydroxide inthe pastes, rutile TiO2 powder was added as internalstandard with weight ratio of 1:5 to the pastes. Rutilediffraction peaks at d=3.24, 2.48 and 1.68 A were usedas reference standard, whereas the calcium hydroxideweight fraction wref was determined from diffractionpeaks at d=4.89, 2.62 and 1.79 A. The integratedintensity data of all patterns containing o/c higher than0.2 were corrected using a calibration curve obtainedcollecting XRD data from binary mixtures of anhy-drous OPC and organoclay for weight ratios 50:50,70:30 and 90:10. In addition, since the peak at d=2.62A of calcium hydroxide is not well resolved because of apartial overlapping of other peaks, the apparent inten-sity of the peak was corrected for the superposition byapplying line profile analysis.The QXRD results, showing the weight ratios wp/wref

of Ca(OH)2 in the cement–clay pastes to Ca(OH)2 in thereference OPC paste (sample R), after 35 and 70 dayscuring, are shown in Table 2. The highest value of the

ratio wp/wref was found for paste C, with wp/wref=1.8after either 35 and 70 days curing. The lowest values ofthe ratios wp/wref were found for pastes A and C00, withwp/wref=1.1 and 1.0 after 35 days curing, and wp/wref=1.3 and 1.0 after 70 days curing. The effect on thehydration reactions by varying the mix compositionscannot be fully explained by considering the nominal w/cratios. Indeed it is considered that in the wet clay usedfor the cement pastes preparation, the amount of waterwas 55 wt.%. The ‘‘true’’ w0/c ratios, calculated by addingthe water content of the wet organoclay to the watermixed with OPC, are also shown in Table 2. Moreover,the nominal clay to cement ratios have to be modifiedtaking into account that the wet clay contained organo-clay, water and 2-CA. The amount of 2-CA was 34 wt.%of the dry clay. Consequently, the nominal clay tocement ratios have been multiplied by 0.45 (to subtractthe water) and then by 0.75 (to subtract the 2-CA). Theresults of these calculations are listed in Table 2 as o0/cand o00/c, respectively. The QXRD analysis showed thatthe most important contribution to the formation ofcalcium hydroxide is the total amount of water in thecement paste. As the weight ratio w0/c was increased,the weight of Ca(OH)2 increased, although the resultssuggest that, when the w0/c ratio is about the same(w0/c=0.73 and 0.70 for pastes B and C00, respectively),a second parameter affects the hydration reactions. Theleaching tests suggested that for all cement–clay pastessome of the 2-CA was not firmly sorbed by the organo-clay, and the amount of 2-CA dispersed in the cementmatrix increased with the w0/c weight ratios. In Table 2the weight ratios a0/c of the amine leached into thecement matrix to cement after 35 and 70 days curing aregiven. The amount of 2-CA dispersed in paste C00 isabout twice as high as that dispersed in paste B and, aswe reported in a previous work (Natali Sora et al.,2002), 2-CA has a retarding effect on the hydrationreactions of cement pastes. However, it has to be noted

Table 2

Composition of mature cement-polluted organoclay pastes; for comparison initial true water content to cement ratio is reported

Pastes
w0/ca o0/cb o00/cc a0/cd wp/wrefe a0/cd wp/wref
e

After 35 days curing
After 70 days curing
R
0.40 – – – 1.0 – 1.0
A
0.51 0.090 0.067 0.003 1.1 0.003 1.3
B
0.73 0.27 0.202 0.012 1.3 0.014 1.3
C
0.95 0.45 0.337 0.027 1.8 0.032 1.8
C0
0.80 0.45 0.337 0.018 1.2 0.026 1.4
C00
0.70 0.45 0.337 0.020 1.0 0.029 1.0
a ‘‘True’’ water to cement weight ratios, w0 indicates the total amount of water in the paste and it is calculated by adding the water content of the

wet organoclay to the water mixed with OPC.b Organoclay to cement weight ratios, o0 indicates the organoclay without water.c ‘‘True’’ organoclay to cement weight ratios, o00 indicates the dry organoclay without water and 2-CA.d 2-CA to cement weight ratios, a0 indicates the amount of 2-CA desorbed from organoclay and dispersed in the cement matrix.e Weight ratios of calcium hydroxide in the clay–cement pastes (wp) to calcium hydroxide in the reference OPC paste (wref) from QXRD

calculations.


that the retarding effect of 2-CA is only evident in pasteswith the same water/cement ratios; indeed, here, apartfrom C00, cement–clay pastes hydrate faster than thereference sample (which has a small w0/c ratio),although some 2-CA is not firmly bound.

3.3. Microstructure

Fig. 4 shows backscattered SEM images of 70 day-oldcement–clay pastes. Backscattered electron imagingcombined with X-ray microanalysis give the followinginformation: light grey areas, in which the Ca/Si atomicratio ranges from 2 to 3, consist of unreacted b-Ca2SiO4

and Ca3SiO5 phases. The grey matrix mainly consists ofhydration products and contains a small amount ofchlorine. Dark grey areas, lacking in reference OPCsamples, have high contents of Si, Al, C and O. Thisclearly is the clay phase, in which also a significantamount of chlorine is found. Generally, as the claycontent of the pastes was increased, there was betterinterspersion of the clay in the pastes and the formationof smaller aggregates. This evidence may be due to thehand mixing of the pastes. Porosity appears larger insamples with greater o0/c ratios, where the increment isprobably related to the effect of clay-bound water ratherthan to the clay itself.

3.4. Mineralogy/degree of hydration

In addition to QXRD, pastes hydration was alsoinvestigated by means of thermogravimetric analyses. Inhydrating cement pastes it is usual practice to speak offree (or unbound) and bound water: the former is thewater that has not yet reacted with cement and the latterthe water that has already reacted. The water which canbe removed by oven-, vacuum-, or freeze-drying (Odler,1998) is conventionally defined as ‘‘free water’’, whilethe remaining water content is called ‘‘bound water’’,and the amount of both types clearly depends on theway used for drying the sample. The removal of freewater from the paste is necessary to stop the hydrationand make the samples suitable for other analyses; in thepresent work, the oven-drying at 85 �C in uncontrolledatmosphere was adopted, even though it is acknowl-edged that this procedure generates changes in thecapillary porosity domain (Galle, 2001).Results concerning free water, as determined by oven

drying, and bound water (by TGA LOI, properly cor-rected) are reported in Fig. 5 for the five series of stabi-lised samples, in addition to the reference cement paste,at two different curing times, i.e. 35 and 70 days,respectively. The amount of free water decreases withageing and the opposite is true for bound water, eventhough a minor discrepancy was detected in sample Band C; in any case, in terms of hydration no dramaticchange can be observed between 35 and 70 days. It is

Fig. 3. XRD patterns of 70-day old cement–clay pastes A, B, C, C0,

C00 and R.


more interesting to compare the hydration of stabilisedwastes with that of the reference paste. In analogy withQXRD analysis, bound water ratios between hardenedsamples and reference (wb/wref) are reported in Table 3(note that wref in QXRD calculations, Table 2, refers tothe portlandite amount in sample R, while in TGAanalyses bound water amounts are indicated with thesame symbol, i.e. wref); these results agree well with

microstructural data (Table 2), confirming the hypoth-eses above put forward.

4. Conclusions

Management of hazardous liquid organics via S/Sprocesses is still a challenge, because the cheapest

Fig. 4. SEM micrographs of 70-day old cement–clay pastes A, B, C, C0, C00 and R.


solidifying agents, i.e. hydraulic binders, are unable toeffectively immobilize these substances. Indeed, organicsinterfere in a detrimental way with the cement hydrationprocess, and they are not firmly bound into the

inorganic siliceous matrix of cement pastes. A viableroute to overcome this drawback seems to pre-sorbthem onto proper materials, such as organoclays. How-ever, even though the results of the waste sorption bythe organoclay appear promising, some issues have tobe further investigated.The portion of 2-CA more easily removed during the

leach test could be the more loosely bound fraction, i.e.bound on the external surface of the clay. Given theweakly basic character of 2-CA, it could be attracted bythe acidic sites on the surface of the organoclays, buteasily removed by water. In addition, the fact that,when admixed with the cement pastes the leached 2-CAamount increases as the water content becomes greater,could be related to the alkaline properties of the cementmatrix. Experiments to assess the behaviour of the

Fig. 5. Free (upper) and bound (lower) water in grams per gram of cement for the six series of samples at 35 and 70 days of ageing; for comparison,

total initial content of water is reported in the upper diagram.

Table 3

Ratio between bound water in polluted samples (wb) and bound water

in the reference sample (R) (wref) calculated by TGA measurements

Sample
wb/wref after
35 days of curing

wb/wref after

70 days of curing

A
1.16 1.15
B
1.66 1.38
C
2.00 1.80
C0
1.04 1.20
C00
1.05 1.10

waste contaminated organoclay with pH variation areongoing.

Acknowledgements

The research was supported from the Ministerodell’Universita e della Ricerca Scientifica through theproject COFIN 1999–2000 (national co-ordinatorProfessor Donatella Botta) and the project COFIN2001–2002 (national co-ordinator Professor GilbertoArtioli). The authors wish to acknowledge LaviosaChimica Mineraria for providing organoclays.

References

Abd El Wahed, M.G., 1991. Electrical conductivity of Portland

cement admixed with some azodyes. J. Mater. Sci. Lett. 10, 1349–

1351.

Andac, M., Glasser, F.P., 1998. The effect of test conditions on the

leaching of stabilised MSWI-fly ash in Portland cement. Waste

Manage. 18, 309–319.

Boyd, S.A., Shaobai, S., Lee, J.F., 1988. Pentachlorophenol sorption

by organo-clays. Clays Clay Miner. 36 (2), 125–130.

Brown, D., Laboureur, P., 1983. The degradation of dyestuffs: Part I.

Primary biodegradation under anaerobic conditions. Chemosphere

12, 397–404.

Chandra, S., Flodin, P., 1987. Interactions of polymers and organic

admixtures on Portland cement hydration. Cem. Concr. Res. 17,

875–890.

Copeland, L.E., Bragg, R.H., 1958. Quantitative X-ray diffraction

analysis. Anal. Chem. 30, 196–201.

Diet, J.-N., Moszkowicz, P., Sorrentino, D., 1998. Behaviour of

ordinary Portland cement during the stabilization/solidification of

synthetic heavy metal sludge: macroscopic and microscopic aspects.

Waste Manage. 18, 17–24.

Dotelli, G., Botta, D., Natali Sora, I., Pelosato, R., 2001. Application

of organophilic clays to the hazardous waste disposal: sorption of 2-

chloroaniline. In: White, T., Sun, D. (Eds.), Advances in Environ-

mental Materials, vol. I: Pollution Control Materials, Symposium

Proceedings ICMAT 2001. Materials Research Society, Singapore,

pp. 145–152.

Edmeades, R.M., Hewlett, P.C., 1998. Cement admixtures. In: Hew-

lett, P.C. (Ed.), Lea’s Chemistry of Cement and Concrete, fourth ed.

Arnold, London, pp. 837–901.

Galle, G., 2001. Effect of drying on cement-based materials pore stru-

cutre as identified by mercury intrusion porosimetry. A comparative

study between oven-, vacuum-, and freeze-drying. Cem. Concr. Res.

31, 1467–1477.

Hamilton, W.P., Bowers, A.R., 1997. Determination of acute Hg

emissions from solidified/stabilized cement waste forms. Waste

Manage. 17, 25–32.

Hills, C.D., Sweeney, R.E.H., Buenfeld, N.R., 1999. Microstructural

study of carbonated cement-solidified synthetic heavy metal waste.

Waste Manage. 19, 325–331.

Jaynes, W.F., Vance, G.F., 1999. Sorption of benzene, toluene, ethyl-

benzene, and xylene (BTEX) compounds by hectorite clays exchanged

with aromatic organic cations. Clays Clay Miner. 47 (3), 358–365.

Krumm, S., 1997. WINFIT beta Release 1.2.1.

Lee, J.F., Mortland, M.M., Chiou, C.T., Kile, D.E., Boyd, S.A., 1990.

Adsorption of benzene, toluene and xylene by two tetra-

methylammonium-smectite having different charge densities. Clays

Clay Miner. 38 (2), 113–120.

Lo, I.M., 1996. C, Solidification/stabilization of phenolic waste using

organic-clay complex. J. Environ. Eng. 122 (9), 850–855.

Lo, I.M.-C., Mak, R.K.M., Lee, S.C.H., 1997. Modified clays for

waste containment and pollutant attenuation. J. Environ. Eng. 123

(1), 25–32.

Montgomery, D.M., Sollars, C.J., Perry, R., Tarling, S.E., Barnes, P.,

Henderson, E., 1991. Treatment of organic-contaminated industrial

wastes using cement-based stabilization/solidification—II. Micro-

structural analysis of the organophilic clay as a pre-solidification

adsorbent. Waste Manage. Res. 9, 113–125.

Natali Sora, I., Pelosato, R., Botta, D., Dotelli, G., 2002. Chemistry

and microstructure of cement pastes admixed with organic liquids.

J. Europ. Ceram. Soc. 22, 1463–1473.

Odler, I., 1998. Hydration, setting and hardening of Portland cement.

In: Hewlett, P.C. (Ed.), Lea’s Chemistry of Cement and Concrete,

fourth ed. Arnold, London, pp. 240–297.

Osmanlioglu, A.E., 2002. Immobilization of radioactive waste by

cementation with purified kaolin clay. Waste Manage. 22, 481–483.

Ouki, S.K., Hills, C.D., 2002. Microstructure of Portland cement

pastes containing metal nitrate salts. Waste Manage. 22, 147–151.

Pagga, U., Brown, D., 1986. The degradation of dyestuffs: Part II.

Behaviour of dyestuffs in aerobic biodegradation tests. Chemo-

sphere 15, 479–491.

Pollard, S.J.T., Montgomery, D.M., Sollars, C.J., Perry, R., 1991.

Organic compounds in the cement-based stabilization/solidification

of hazardous mixed wastes—mechanistic and process considera-

tions. J. Hazard. Mater. 28, 313–327.

Sheng, G., Xu, S., Boyd, S.A., 1996. Cosorption of organic con-

taminants from water by hexadecyltrimethylammonium-exchanged

clays. Water Res. 30 (6), 1483–1489.

Smith, J.A., Jaffe, P.R., Chiou, C.T., 1990. Effect of ten quaternary

ammonium cations on tetrachloromethane sorption to clay from

water. Environ. Sci. Technol. 24, 1167–1172.

UNICEN 8798, December 1986. Radioactive waste solidification

products—Long term leach test (in Italian). UNI Ente Nazionale

Italiano di Unificazione, Milano, Italy.

Wilding, C.R., Walter, A., Double, D.D., 1984. A classification of

inorganic and organic admixtures by conduction calorimetry. Cem.

Concr. Res. 14, 185–194.

Wiles, C.C., 1987. A Review of stabilization/solidification technology.

J. Hazard. Mater. 14, 5–21.


Documents

Cement–clay pastes for stabilization/solidification of 2-chloroaniline