water research 43 (2009) 2605–2614

Available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/watres

Precipitation of heavy metals from wastewater using

simulated flue gas: Sequent additions of fly ash,
lime and carbon dioxide

Quanyuan Chena,*, Zhou Luoa, Colin Hillsb, Gang Xuea, Mark Tyrerc
a
School of Environmental Science and Engineering, Donghua University, 2999 Renmin Road, Shanghai 201620, PR China
b
Centre for Contaminated Land Remediation, School of Science, University of Greenwich, Chatham Maritime, Kent ME4 4TB, UK
c
Department of Materials, Imperial College of Science, Technology and Medicine, London SW7 4AZ, UK

article info abstract

Article history: Lime is a preferred precipitant for the removal of heavy metals from industrial wastewater
Received 24 October 2008 due to its relatively low cost. To reduce heavy metal concentration to an acceptable level
Received in revised form for discharge, in this work, fly ash was added as a seed material to enhance lime precip-
25 January 2009 itation and the suspension was exposed to CO2 gas. The fly ash–lime-carbonation treat-
Accepted 1 March 2009 ment increased the particle size of the precipitate and significantly improved
Published online 17 March 2009 sedimentation of sludge and the efficiency of heavy metal removal. The residual concen-
trations of chromium, copper, lead and zinc in effluents can be reduced to (mg L1) 0.08,
Keywords: 0.14, 0.03 and 0.45, respectively. Examination of the precipitates by XRD and thermal
Precipitation analysis techniques showed that calcium–heavy metal double hydroxides and carbonates
Double hydroxide were present. The precipitate agglomerated and hardened naturally, facilitating disposal
Hydroxyl carbonate without the need for additional solidification/stabilization measures prior to landfill. It is
Carbonation suggested that fly ash, lime and CO2, captured directly from flue gas, may have potential as
Flue gas a method for wastewater treatment. This method could allow the ex-situ sequestration of
Carbon dioxide CO2, particularly where flue-gas derived CO2 is available near wastewater treatment
facilities.
ª 2009 Elsevier Ltd. All rights reserved.

1. Introduction et al., 1996; Lazaridis et al., 2001; Garcia-Sanchez and Alvarez-

Industrial activities give rise to large quantities of wastewater
containing toxic heavy metals. In contrast to organic pollut-
ants in wastewater, heavy metals can only be removed by
separation or converted to a chemically inert state (Char-
erntanyarak, 1999; Alvarez-Ayuso et al., 2003; Kurniawan
et al., 2006). The removal of heavy metals from wastewaters
can be achieved by, for example, chemical precipitation, ion
exchange, solvent extraction, ion flotation, adsorption and
membrane filtration (Tuenay and Kabdasli, 1994; Baltpurvins
Ayuso, 2002; Doyle and Liu, 2003; Tuenay, 2003; Dabrowski
et al., 2004; Wingenfelder et al., 2005; Khedr, 2008; Sudilovskiy
et al., 2008). Of these, chemical precipitation is the most
widely used method due to its simplicity of use (Li et al., 2003;
Duan and Gregory, 2003). In this approach, the dissolved metal
ions are converted to the insoluble solid phases via a chemical
reaction with a precipitant, for example, alkali or sulfide. The
resultant precipitate is then separated from the water by
sedimentation and/or filtration or flotation (Matis et al., 2004;
Zamboulis et al., 2004).

* Corresponding author. Tel.: þ86 21 6779 2540; fax: þ86 21 6779 2552.
E-mail address: qychen@dhu.edu.cn (Q. Chen).
0043-1354/$ – see front matter ª 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.watres.2009.03.007
2606 water research 43 (2009) 2605–2614
Lime (CaO) is a preferred precipitant for the removal of
heavy metals from industrial wastewater, but a comparatively
high dosage is required and heavy metals may not be reduced
to an acceptable level for discharge due to poor inadequate
settling and the dissolution of precipitates (Tadesse et al.,
2006). The precipitates of amphoteric metals, for example,
zinc and lead, tend to re-dissolve as the pH changes beyond
the optimal range. Consequently, pH must be strictly
controlled and a fractional precipitation approach is usually
needed for the treatment of complex wastewaters containing
different heavy metals.
The efficiency of sludge/water separation by sedimenta-
tion may be reduced by the presence of colloidal particles,
even though they may be suitably filterable with a 0.45 mm
membrane filter in laboratory. The solution conditions,
including super-saturation, temperature, reagent concentra-
tions and pH, influence the properties of the precipitates such
as polymorphism, morphology, particle size and solubility.
It is reported that the slaking of CaO in the presence of Na2SO4
gives Ca(OH)2 crystal as hexagonal plates between 5 and 20 mm
in size, whereas in the presence of NaCl or NaOH, Ca(OH)2
crystals between size of 2 and 5 mm are formed, which
are slightly larger than those formed in the absence of
admixtures (Glasson, 1961; Ritchie and Xu, 1990; Giles et al.,
1993; Xu et al., 1997, 1998). As separation of lime precipitates
from treated wastewater is difficult to achieve at a commer-
cial scale, methods that change the particle size distribution of
the precipitate may provide some advantages (Marani et al.,
1995; Lim et al., 2002; Lee et al., 2005). Coagulation and floc-
culation are common methods to destabilize colloidal parti-
cles and promote the growth of larger, more easily removed
flocs. The efficacy of coagulating agents depends, for example,
upon effluent pH and the nature of any organic compounds
present (Licsko, 1997; Juang and Shiau, 2000; Duan and
Gregory, 2003). The additional cost and resultant high residual
concentration of sulfate and aluminium ions can negate this
approach (Li et al., 2003). For precipitates with low solubility
products, such as heavy metal hydroxides, very small parti-
cles normally dominate because the level of super-saturation
at the feed points is difficult to control by micro-mixing.
Industrial practice is, therefore, to induce seeding to promote
crystallization as an effective measure for increasing the
particle size.
Fly ash is particulate material produced from the combus-
tion of coal in thermoelectric power plants. It gives rise to a pH
of 10–13 when mixed with water at a solid/liquid ratio of
around 10. Hence, it can be reasonably expected that metal ions
can be removed from aqueous solutions by precipitation and
adsorption onto fly ash particles. According to a number of
authors (Cho et al., 2005; Wang and Wu, 2006; Alinnor, 2007;
Ngah and Hanafiah, 2008), addition rates in excess of 20 g L1
are usually required because of its inherently low adsorption
capacity, resulting in significant separation and disposal costs
when the fly ash becomes ‘spent’. There is a paucity of infor-
mation on the maintenance of desirable geo-technical prop-
erties whilst ensuring adequate treatment of contaminants,
especially within a landfill scenario (Holt and Freer-Hewish,
1996). It also appears that fly ash has not been used commercial
adsorbent for wastewater treatment, but may have potential as
an ideal seeding material for lime precipitation.
Carbon dioxide can neutralize the alkaline nature of lime
and reduce the solubility of some heavy metal compounds.
Carbonation is used to reduce the leaching of hazardous
substances and to induce setting and strength development in
cement-based solidified/stabilized waste forms (Hills et al.,
1999; Lee et al., 2005). It is well known that carbonation has
been used for preparing potable water (Crittenden et al., 2005).
The application of carbonation to the treatment of wastewater
containing oxyanions, has been recently reported by Montes-
Hernandez et al. (2008, 2009).
In the present work, a small addition of fly ash (as a seed
material) and lime, followed by a carbonation step, was used
to remove heavy metals from solutions. The agglomeration of
the precipitates during the treatment was investigated and
the resulting sludge was then characterized by XRD and DTA/
TG techniques.
2. Experimental
2.1. Materials
Solutions (synthetic effluents) containing Cu(II), Cr(III), Pb(II)
and Zn(II) were prepared from their respective standard
reagent grade metal nitrate salts (BCD Ltd.). Each metal salt,
Cu(NO3)2$3H2O, Zn(NO3)2$6H2O, Cr(NO3)3$9H2O and Pb(NO3)2,
was dissolved in de-ionised water at a concentration of
100 mg L1 for precipitation studies and at a concentration of
50 g L1 for the investigation of pH (as heavy metal ion,
assuming ideal stoichiometry). The pH of heavy metal nitrate
solutions was in the range of 2.0–2.7.
Fly ash was a heterogeneous material consisting largely of
small aluminous and siliceous glass spheres. The majority
of spheres (>90%) were less than 45 mm and mean particle
diameter was 37 mm (specific surface area: 1.2 m2 g1). The
main constituents of fly ash were (%): SiO2 42.5, Al2O3 28.3,
Fe2O3 11.1, CaO 7.5, MgO 2.2, TiO2 1.9, K2O 0.4, Na2O 0.4, and
SO3 0.3. The mass lost on ignition at 1000  C was 3.5%. Fly
ash was prepared for use by oven drying at 110  C for 2 h.
The calcium oxide used was of 98% purity (Aldrich Chemical
Company).
2.2. Methods
2.2.1. Precipitation
The precipitation of heavy metals from the synthetic effluents
involved batch experiments at room temperature (20  1  C) in
sealed 500 mL polyethylene vessels. Fly ash was added at
a rate of 0.1 g to 200 mL of the heavy metal solutions and then
lime was added at a desired amount. After the addition of
lime, the suspensions were agitated at 200 rpm for 3 min using
a magnetic stirrer, followed by gentle stirring at 40 rpm for
7 min to promote coagulation and agglomeration. The pH was
constantly monitored using a Philips DW9418 pH meter with
a glass electrode. After mixing, the suspensions were allowed
to settle for 30 min, and the supernatant (20 mL) from each
vessel was collected and acidified to <pH 2 by adding 5 mL
of nitric acid, filtered through 0.45 mm Whatman filters
and analyzed by ICP-AES (Perkin–Elmer, Optima 4300DV).
The turbidity of the supernatants was determined at 450 nm
 water research 43 (2009) 2605–2614 2607

by an HI 93703 microprocessor turbidity meter calibrated

against a formazin standard. 3. Results and discussion
Carbonation of the suspensions (synthetic wastewater
with fly ash and lime) involved stirring with a magnetic stirrer 3.1. Heavy metal removals
in a 99.9% carbon dioxide atmosphere maintained at a pres-
sure of 0.2 MPa. The suspensions were stirred at 40 rpm until A fly ash addition of 20 g L1 can be used to promote the
a desired pH was attained and then allowed to settle for adsorption of heavy metals (Wang and Wu, 2006; Alinnor,
30 min, prior to sampling and preparation for analysis by ICP- 2007; Ngah and Hanafiah, 2008). In this work, a lower ‘seed’
AES as described above. dose of 500 mg L1 fly ash was used, to promote a minimal
amount of sludge production. With fly ash alone the removal
2.2.2. Sludge volume of heavy metals was below 10%, but with incremental
The volume of the decanted sludge was estimated by the additions of lime up to 900 mg L1, metals removal improved
volumetric method using Imhoff cones. After 30 min of dramatically (see Fig. 1) until it steadied at 97.14%–98.54%. It
settling, the volume of sludge produced was determined by can be seen that the concentration of Pb and Cr exceeded the
directly in mL per L of wastewater treated. The solids were prescribed limits (1.0 mg L1, and 1.5 mg L1, respectively)
then filtered, dried at 105  C for 24 h and weighed. stipulated in The Integrated Wastewater Discharge Standard
of China (GB8978-1996), indicating that other physico-
chemical treatment methods were required for compliance
2.2.3. Leaching
with this discharge standard.
To evaluate the performance of precipitates in landfill envi-
The removal of metal ions under these experimental
ronment, the lime precipitate sludge was leach tested
conditions was dominated by precipitation. Fig. 2 shows the
followed the German procedure (DIN38414-S4). The vacuum
dependence of heavy metals precipitation on suspension pH,
oven dried sludge of 10 g (crushed to <1 mm) was rotated on
which was, where necessary, regulated by H2SO4 additions, as
a roller at 30 rpm with 100 mL of distilled water for 24 h at an
lime dosage was 900 mg L1. The efficiency of metal precipi-
ambient temperature of 20  C. The vessels were centrifuged
tation was high in the broad range pH 6–12, with the optimal
and the supernatant liquids filtered through a 0.45 formazin
value being between pHs 8 and 10. Of the four heavy metal
membrane filter prior to acidification and analysis by ICP-AES.
ions of interest, the removal of copper was the greatest
Leaching experiments were performed in duplicate and all
decreasing in the sequence: copper, lead, chromium and zinc.
analyses were conducted in triplicate, and reported as an
As shown in Fig. 2, copper was removed at a broader pH range
average of the values obtained.
when fly ash and lime were used together. At pH 4, 70% of
copper was precipitated, increasing to 99.31% at pH 6.
2.2.4. X-ray powder diffraction (XRD)
The residual Cu concentration was 0.69 mg L1, which met the
A Siemens D500 diffractometer and Kristalloflex 810 gener-
discharge limit for Cu (0.5–2.0 mg L1, GB8978-1996). The
ator (CuK-alpha radiation, 40 kV, 40 mA) were used to identify
precipitation of zinc increased rapidly with pH and 97% was
the crystalline products in the dried sludge, ground to
removed between pHs 8 and 10. For lead, 98% removal
<30 mm. Diffractograms were obtained between 5 and 40 2q
occurred between pHs 7 and 11. The optimum pH range for
using Diffplus, and Bruker/AXS EVA software utilizing the
chromium precipitation using fly ash and lime was 7–10.
current version of the international powder data file (ICDD-
In the heavy metal (50 g L1)/fly ash (500 mg L1)/lime
JCPDS).
(900 mg L1) suspensions, pH increased with time and at
different rates in the sequence: Cu > Cr > Pb > Zn (see Fig. 3),
2.2.5. Thermal analysis (DTA/TG)
reflecting the interaction of the various species in solution.
A Stanton Redcroft STA 780 simultaneous thermal analyzer
According to the pH variation rate, heavy metals retarded CaO
was used to conduct DTA/TG analyses on the precipitates.
Vacuum-dried powder samples (<30 mm, 20 mg) at a temper-
ature of 60  C for 72 h were placed in a rhodium–platinum
(Pt87–Rh13) crucible (5.8 mm dia, 4 mm high). Samples were
100
subject to a heating rate of 10  C min1 under flowing nitrogen
90
(40 cm3 min1) from 30 to 900  C. The calcium hydroxide
and calcium carbonate contents were determined from 80

following equations: 70 Pb
Removal, %

60 Cr
CH (%) ¼ WLCH (%)  MWCH/MWH (1) 50 Zn
40
Cu
30 fly ash 500 mg L-1
CC (%) ¼ WLCC (%)  MWCC/MWC (2)
20

where CH (%) and CC (%) are the content of Ca(OH)2 and 10

CaCO3, WLCH (%) and WLCC (%) are the weight loss occurred 0
0 100 300 500 700 900 1100 1300
during the decomposition of portlandite and calcium
lime dose (mg/L)
carbonate, MWCH, MWH, MWC and MWCC are the molar weight
of portlandite, water, carbon dioxide and calcium carbonate. Fig. 1 – Lime dosage and heavy metal removals.
2608 water research 43 (2009) 2605–2614

100 100

90
90
80

Removal, %
80
Removal, %

70
70 Pb Cr
60
Zn Cu Pb Cr
60 50
Cu Zn
40
50 fly ash 500 mg L-1, lime 900 mg L-1
30
40 2 4 6 8 10
4 5 6 7 8 9 10 11 12 Time, min
pH
Fig. 4 – Precipitation kinetics of heavy metals.
Fig. 2 – Heavy metal precipitation on suspension pH.

slaking in the decreasing order: Cu < Cr < Pb < Zn. Heavy
metal precipitation on the surfaces of lime influenced the
slaking process. According to Ritchie and Xu (1990) and Giles
et al. (1993), the slaking of calcium oxide and the dissolution of
calcium hydroxide are both controlled by diffusion of calcium
or hydroxide ions away from particle surfaces. Hydroxide ions
have a retarding effect on CaO hydration due to the common
ion effect. In contrast, the acidic conditions can accelerate the
hydration of lime and the diffusion of other ions towards the
surface of calcium oxide is also found to influence the slaking
process. Xu et al. (1997, 1998) reported that the reaction of
calcium oxide with aluminium ions was initially controlled by
diffusion of aluminium to the calcium oxide surface, forming
a film of calcium aluminate that blocked further reactions
before CaO was eventually consumed. They observed
a doubling of the rate of CaO slaking in the presence of 1–3%
chloride, whereas with 0.1–3% sulfate or carbonate the rate of
hydration of calcium hydroxide decreased.
The kinetics of heavy metal removal at an initial metals
concentration of 100 mg L1 using fly ash (500 mg L1) and
calcium oxide (700 mg L1) was also investigated (Fig. 4). These
suspensions were magnetically stirred at 40 rpm with
precipitation completion being reached within 10 min for all
14
12
10
8 Cr
pH
the solutions. The change in solution pH observed correlated
with the sequence of heavy metal removal.
To enhance the precipitation of lime, the stirred suspen-
sions were exposed to CO2 with a purity of 99.9% at 0.2 MPa.
The suspensions had an initial metal concentration of
100 mg L1, and contained 500 mg L1 of fly ash and 900 mg L1
of lime. On contact with CO2, the suspension pH values were
in the range 10–6. As is well known, the pH of water at 18  C in
equilibrium with atmospheric CO2 is 5.7, whereas in equilib-
rium with 100 vol.% of CO2, the pH can reach 3.9 (Stumm and
Morgan, 1995). In these experiments the pHs were higher
because of the dissolution of alkaline components from fly
ash and residual lime. Carbonation increased heavy metal
removal significantly as shown in Fig. 5. Lime precipitation
with the addition of fly ash followed by carbonation was
capable of reducing metals concentrations to meet the regu-
latory requirements. The addition of CO2 reduced residual Cu
to 0.14 mg L1, far below the Chinese discharge limit (0.5–
2.0 mg L1, GB8978-1996) and the USEPA enforceable limit of
0.25 mg L1 (Environmental Protection Agency, 2002, 2004;
Crittenden et al., 2005). The residual metals levels: Cr
0.08 mg L1, Pb 0.03 mg L1, and Zn 0.45 mg L1, satisfied the
following regulatory limits: Cr 1.5 mg L1, Pb 1.0 mg L1, and
Zn 2.0 mg L1 (GB8978-1996).
After a 10 min of agitation, the turbidity of the settling
suspensions was measured and the results are given in Fig. 6.
The lime precipitates obtained without fly ash and carbon-
ation were too small to naturally settle and the turbidity of the
suspension remained as high as 175 NTU, after 30 min of
sedimentation. In contrast, the suspensions containing fly ash
and carbonation settled rapidly and met the Chinese effluent
standard of 70 NTU.

Cu
6
Pb
4 3.2. Sludge characterisation
Zn

2 Control 3.2.1. Sludge volume

As is shown in Table 1, treatment by carbonation resulted in
0 lower sludge production. The amount recorded was 93–
0 3 5 10 20 30 40 50 60
127 mL L1 (depending on the heavy metal), compared to 210–
Time, min
288 mL L1 for lime only, showing that with fly ash and CO2,
Fig. 3 – Suspension pH with time. the treated wastewater was more easily de-watered.
 water research 43 (2009) 2605–2614 2609

250
a 0.7
lime
0.6 200 FA+lime

Turbility, NTU
FA+lime+CO2
0.5
150
Cu, mg L-1

Hydration
0.4
Carbonation 100
0.3

0.2 50

0.1 0
5 10 20 30 40 50 60
0
7 8 9 10 11 12 Time (min)
pH
Fig. 6 – Turbidity of suspensions during settling.

5
b
Hydration metals from the carbonated precipitates was much lower than
4
that from the non-carbonated precipitates.
Carbonation
Pb, mg L-1

3
3.2.3. Phase identification
2
3.2.3.1. Non-carbonated precipitates. The crystalline phases
1 identified in the non-carbonated precipitates are shown in
Fig. 7. Calcium–heavy metal double hydroxides, such as
0 2Ca(OH)24Cu(OH)2$H2O, CaZn2(OH)6$2H2O and Ca3Cr2(OH)12
7 8 9 10 11 12 were formed during the precipitation process. As may be
pH expected, the heavy metal hydroxides (Pb(OH)2, Cu(OH)2,
Zn(OH)2) were also observed. It should be noted that Cr(OH)3 is
c 14
usually amorphous and cannot be detected by XRD (Chen et al.,
12 Hydration 2007; Chen et al., 2009). The presence of calcite can be attrib-
uted to atmospheric carbonation during sample preparation.
10 Carbonation
The DTA curves and simultaneous TG curves obtained
Zn, mg L-1

8 from non-carbonated precipitates are given in Fig. 8. In the

6 control sample, an endotherm from portlandite dehydrox-
ylation occurred at 420–500  C, whereas the endotherm at 680–
4
750  C represented the decomposition of calcite. Corre-
2 spondingly, two mass losses can be seen in the TG curve, and
it was estimated that the contents of portlandite and calcite
0
8 9 10 11 12 were 61.7% and 27.1%, respectively.
pH In the Cr precipitate, there were five major endothermic
peaks and five notable mass loss steps at 100–120  C, 280–
d 10 310  C, 350–430  C, 450–600  C and 650–800  C, respectively.
The first endothermic peak can be attributed to the evapora-
8 tion of water, for example, the structural water in double
hydroxide such as Ca2Cr(OH)7$3H2O. The second peak could
Cr, mg L-1

6 Hydration
be due to the dehydroxylation of calcium–chromium double
Carbonation hydroxide. The mass change between 350 and 600  C was
4
associated with the dehydroxylation of calcium–chromium

0 Table 1 – Sludge characteristics.

6 7 8 9 10 11 12
Metals Sludge amount, mL L1 Leachate concentration,
pH
mg L1
Fig. 5 – Heavy metal concentrations and precipitating pH.
Lime Lime þ fly Lime Lime þ fly
ash þ CO2 ash þ CO2

Cr 282 116 0.59 0.44

Cu 210 93 2.65 1.06
3.2.2. Leaching Pb 245 127 0.55 0.39
The results obtained from the leaching of sludge are also Zn 288 125 6.72 3.38
presented in Table 1. It can be seen that the release of heavy
2610 water research 43 (2009) 2605–2614
Fig. 7 – Diffractograms of non-carbonated heavy metal precipitates. (C: calcite, P: portlandite, CuH: 2Ca(OH)24Cu(OH)2$H2O,
CaZn: CaZn2(OH)6$2H2O, CaCr: Ca2Cr(OH)7$3H2O, CaCr2: Ca3Cr2(OH)12).
double hydroxide (Ca3Cr2(OH)12, Ca2Cr(OH)7$3H2O) and por-
tlandite. As well known, portlandite decomposition occurred
in the range of 380–450  C. The last mass change corresponds
to the fifth endothermic peak at 650–800  C and was assigned
to the decomposition of calcium carbonate. The amount of
calcium carbonate present was calculated to be 44.1% by
mass, much higher than the control sample (around 27%),
indicating that the presence of chromium promoted atmo-
spheric carbonation during sample preparation.
The Cu precipitate exhibited five endothermic peaks and
corresponding distinctive mass loss steps. The first endothermic
peaks present can be attributed to the loss of chemical bound
and physical bound water. The double endothermic peaks at
200–400  C were due to the decomposition of copper and calcium
double hydroxide (2Ca(OH)24Cu(OH)2H2O). The strong endo-
therm at around 450  C can be attributed to dehydroxylation of
portlandite, whereas the last endothermic peak represented the
decomposition of calcium carbonate (37.5% by mass).
The DTA/TG curves from the Pb precipitate were very
similar to those of the control sample, except that the
temperature of portlandite dehydration was lower, shifting
from 420–500  C to 390–450  C. This suggested that Pb could
substitute for Ca in portlandite and reduce its thermal stability.
Five major mass loss steps were observed for the Zn
precipitate. The mass loss between 100 and 200  C was
attributed to the evaporation of structural water in calcium
zincate (CaZn2(OH)6$2H2O), whereas the second and third
were attributed to the decomposition of this phase. The fourth
endothermic peak and corresponding mass loss step arose
from the dehydroxylation of portlandite. The last endotherm
was due to the decomposition of calcium carbonate.
3.2.3.2. Carbonated precipitates. The diffractograms obtained
from the carbonated precipitates are given in Fig. 9. Double
hydroxides, including CaZn2(OH)6$2H2O, Ca2Cr(OH)7$3H2O
and 2Ca(OH)24Cu(OH)2$H2O, were detectable, in addition to
PbCO3. The presence of portlandite was probably due to
incomplete carbonation.
The DTA and simultaneous TG curves of carbonated heavy
metal precipitates are presented in Fig. 10. In the carbonated
Cr precipitate, there were four obvious endothermic peaks in
the DTA curve and four corresponding mass loss steps, which
were located at 100–120  C, 250–340  C, 450–540  C and 650–
800  C, respectively. The first endotherm can be attributed to
the loss of chemically and physically bound water. The second
could be due to the dehydration of calcium–chromium double
hydroxide (Ca2Cr(OH)7$3H2O), whereas the third mass loss
step arose from the dehydroxylation of calcium chromium
double hydroxide. The last mass loss step represented the
decomposition of calcium carbonate, which was determined
to be 55.2% by mass.
In the carbonated Cu precipitate, the first endothermic peak
(100–120  C) arose from the water loss of chemical bound water.
The second and third peaks (240–270  C, 440–500  C) confirmed
the presence of copper–calcium double hydroxide (2Ca(OH)24-
Cu(OH)2$H2O). The endotherm at 680–800  C was due to the
decomposition of calcium carbonate (66.8% by mass).
The DTA and TG curves of the carbonated Pb precipitate
were analogous to those of the carbonated control sample. The
temperature range at which calcite decomposed was, however,
much narrower and lower (660–800  C) and the calcium
carbonate content was determined to be 74.6% by mass.
In the carbonated Zn precipitate, three major endothermic
peaks were recorded. The first of the 3 mass loss steps
between 100 and 480  C might be attributed to the dehydration
of calcium–zinc double hydroxide, i.e. calcium zincate (CaZ-
n2(OH)6$2H2O), whereas the mass loss at 480–540  C arose
from its decomposition. The third endotherm was due to the
decomposition of calcium carbonate, which was determined
to be 65.1% by mass.
It was noted that the decomposition temperature range for
calcite in the presence of heavy metals was close (650–750  C),
but was much lower than that of calcite in the control sample.
This reduction in temperature could be attributed to the surface
adsorption of heavy metals or the substitution of heavy metals
for calcium in the carbonate lattice, lowering the thermal
stability of calcite. Garcia-Sanchez and Alvarez-Ayuso (2002)
have investigated heavy metals adsorption on the surface of
calcite and found that heavy metals were retained on calcite
according to the following sequence: Cr(III) 217 mg g1 > Zn(II)
 water research 43 (2009) 2605–2614
Fig. 8 – DTA/TG curves of non-carbonated heavy metal
precipitates.
185 mg g1 > Cd(II) 16.3 mg g1. Chen et al. (2008) observed
adsorption and surface precipitation phenomena of heavy
metals on the composite of silicate gel and calcite.
3.3. Heavy metal speciation and mechanism of
precipitation
The precipitation of heavy metals is dependent on the
hydration and dissolution of fly ash, CaO and CO2 as well as
metal speciation. The hydration of CaO proceeds according to
the following hydrolysis reaction:
CaO þ H2O / Ca2þ þ 2OH (3)
Fly ash is initially sparingly dissolved in water. In the
absence of metal ions it reaches equilibrium in about 30 min,
when a relatively constant concentration of calcium and
silicon ions and a constant pH are established (Wang and Wu,
2006; Alinnor, 2007; Ngah and Hanafiah, 2008). The solid
2611
phases in fly ash, for example, gehlenite, Ca2Al2SiO7, can
slowly release hydroxyl ions according to the following
hydrolysis reaction:
Ca2Al2SiO7 þ 7H2O / H4SiO4 þ 2Al(OH)3 þ 2Ca2þ þ 4OH (4)
In the presence of heavy metal ions, sorption and precipi-
tation phenomena occur. Hydroxyl ions generated by reac-
tions Eqs. (3) and (4) result in the precipitation of heavy metal
hydroxides. The precipitates of amphoteric metals, including
chromium, zinc and lead, tend to re-dissolve as pH changes
beyond the optimal range. As calcium ion concentrations in
the liquid phase change, the co-precipitation of calcium and
heavy metals occurs augmented by the release of, for
example, Al and Si from the fly ash.
The effect of lime on the precipitation of heavy metals is
influenced by the chemistry of the particular metal solutions,
as the distribution of metal species is dependent on both total
metal concentration and the pH of the equilibrium solution.
For Cu(II), the overall formation constants (log b) for
Cu(OH)þ, Cu(OH)2, and Cu(OH)2 4 are 6.3, 11.8, and 16.4,
respectively (Stumm and Morgan, 1995). Copper occurs
predominantly as Cu2þ species up to pH w 6.0. The concen-
tration of various Cu hydroxyl species, including CuOHþ,
[Cu2(OH)2]2þ, Cu(OH)2 (aq) and [Cu(OH)3], increases at
2
pH > 5.0. Other Cu species such as CuNOþ 3 , [Cu(OH)4] and
3þ
[Cu2(OH)] are negligible in solution and do not significantly
change with pH. At pH 6–8, and 8–12, the dominant species of
Cu(II) are Cu2(OH)2þ 2 , and Cu(OH)2 (aq), respectively (Xue et al.,
2009). Considering the alkaline conditions prevailing, most of
these metals might precipitate as copper hydroxides and
calcium–copper double hydroxide (e.g. 2Ca(OH)24-
Cu(OH)2$H2O). In the presence of CO2, in addition to H2CO3,
HCO 3 predominates in carbonate solutions in the pH range
from 6 to 11. Copper precipitates as carbonates and calcium–
copper double hydroxide (2Ca(OH)24Cu(OH)2$H2O).
Zinc occurs predominantly as Zn2þ species at pH <w7.5 and
is expected to form hydroxides at pH 8–10. At pH 7.5–11.5 and
pH > 11.5, the dominant species of Zn(II) are Zn(OH)2 (aq),
and Zn(OH) 3 , respectively (Stumm and Morgan, 1995; Xue
et al., 2009). In addition to Zn hydroxy-complexes, a calcium–
zinc complex hydrated compound (CaZn2(OH)6$2H2O) may
also form at pH 7–12. Zinc hydroxide, a typical amphoteric
hydroxide functions as both an acid and a base:
Zn2þ þ 2OH / Zn(OH)2 / 2Hþ þ ZnO2
2 (5)
2
The hydroxy-complexes Zn(OH)2 (aq), Zn(OH) 3 , Zn(OH)4
3
and Zn(OH)5 can be present in a strong alkaline solution. The
2
presence of Zn species such as ZnNOþ 3 , Zn(OH)4 and
þ
[Zn2(OH)]3 is negligible.
In the presence of CO2, the predominant aqueous Zn
species are di-carbonates Zn(CO3)2 2 and ZnCO3, due to its high
tendency to form tetrahedral complexes. In addition, calcium
zincate precipitation occurs.
For lead, the dissolved species in solution are Pb2þ, PbOHþ,
2þ
Pb(OH)2 (aq), and Pb(OH) 3 . Lead occurs predominantly as Pb
species up to pH w 6. It is slightly different to Cu and Zn, as the
nitrate species PbNOþ 3 is significant at pH < 6.0, after which
PbOHþ is the dominant dissolved Pb(II) species. Small
2612 water research 43 (2009) 2605–2614

Fig. 9 – Diffractograms of carbonated heavy metal precipitates. (C: calcite, P: portlandite, CuH: 2Ca(OH)24Cu(OH)2$H2O, CaZn:
CaZn2(OH)6$2H2O, CaCr: Ca2Cr(OH)7$3H2O).

amounts of other Pb species (Pb(NO3)2 (aq)), Pb4(OH)4þ 4 , and The data obtained from XRD and DTA analyses and Cr(III)
Pb2(OH)þ 3 also occur in solution. At pH 7.5–9, 9–10, 10–11, and speciation indicted that Cr(III) precipitated as a calcium–
pH > 12, the dominant species of Pb(II) are PbOHþ, Pb3(OH)2þ 4 , chromium double hydroxide (Ca2Cr(OH)7$3H2O) and Cr(OH)3,
Pb(OH)2, and Pb(OH) 3 , respectively (Stumm and Morgan, 1995; with maximum precipitation occurring at pH w 9 on addition
Xue et al., 2009). With an increase in pH, Pb can form of Ca(OH)2. In the presence of CO2, pH was reduced,
hydroxides. In the presence of CO2, the lead hydroxo complex hampering the formation of higher soluble Cr(VI) species. The
formation is highly inhibited by the competitive action of the optimum pH for precipitation of Cr(III) was found to be 6–9.5.
bicarbonate ligand. Lead carbonate is the predominant lead
species in the pH range from 6 to 11 as predicted by the solid/
3.4. Environmental implications of this research
solution equilibrium of the Pb(II)–H2O–CO2 system, and lead
carbonate can precipitate at near neutral pH even when air-
The use of CO2, fly ash and lime for the precipitation of heavy
saturated water is used to prepare lead nitrate solutions. The
metals may develop a new method for wastewater treatment
X-ray diffraction analysis identified cerussite (PbCO3), but
by flue gas. This could be an effective way for reducing CO2
hydrocerussite (Pb3(CO3)2(OH)2), suggesting the precipitating
emission. As the anthropogenic emission of CO2 may be
conditions favored the formation of the former.
responsible for the current trend of global warming, the use of
Chromium exists in seven main forms in aqueous solution,
flue gas in wastewater treatment is worth noticing. This
namely Cr3þ (pH < 6), CrOH2þ (pH 2–7.5), Cr2(OH)4þ 2 (pH 3–5),
method excels the conventional method of chemical treat-
Cr3(OH)5þ þ
4 (pH 2.5–8), Cr(OH)2 (pH 4–10), Cr(OH)3 (pH 6–12) and
ment, showing certain prospect of spread and application. It
Cr(OH) 4 (pH > 7.5). At pH 4–6.5, 6.6–8.5, 8.5–9.5 and >9.5, the
could allow the ex-situ sequestration of CO2, particularly,
dominant species of Cr(III) are Cr3(OH)5þ þ
4 , Cr(OH)2 , Cr(OH)3 and
2 where flue-gas derived CO2 is available near wastewater
Cr(OH)4 , respectively. Cr(VI) species such as HCrO

4 , CrO4 ,
 2 treatment facilities. The most likely applications of this
HCr2O7 and Cr2O7 are negligible at pH < 11, as these Cr(VI)
method could be the treatment of metallurgical industry
species are strongly oxidizing (Viguri et al., 2000; Lazaridis
wastewater, where heavy metal concentrations ranging from
et al., 2001). Reduction of these species is accompanied by the
tenths to hundreds of mg L1 and flue gas is readily available.
Hþ consumption as follows (Zouboulis et al., 1995):
Montes-Hernandez et al. (2008, 2009) investigated the
removal of metal ions from synthetic wastewater via the
Cr2O2 þ 
7 þ 14H þ 6e ¼ 2Cr
3þ
þ 7H2O (6) carbonation of calcium hydroxide under moderate pressure
(pCO2 20 bar) and temperature (30  C). The Ca(OH)2 carbon-
ation reaction promoted the removal of selenite (>90%), arse-
3þ
HCrO þ 
4 þ 7H þ 3e ¼ Cr þ 4H2O (7) nate (>78%) and phosphate (z100%) from synthetic solutions
 water research 43 (2009) 2605–2614
Fig. 10 – DTA/TG curves of carbonated heavy metal
precipitates.
by incorporation and/or adsorption into and onto minerals.
Oxyanions may substitute for carbonate ions (CO2 3 ) during
calcite crystal precipitation and growth. It was shown that the
kinetics of CO2 transfer (during calcite formation) in presence
of selenite and phosphate was significantly retarded. Never-
theless, the aqueous carbonation of calcium hydroxide with
compressed CO2 could be an economical and ecological
method to remove various oxyanions as well as to allow the ex-
situ mineral sequestration of CO2; particularly, when the
Ca(OH)2 is available from alkaline solid waste, for example, fly
ash, waste concrete and cement and alkaline paper mill waste.
4. Conclusions
The precipitation of heavy metals with calcium oxide in the
presence of fly ash has been shown to be successful in
reducing the level of soluble heavy metals in solution. The
application of carbonation improved the removal of heavy
metals from the synthetic wastewater over a wide range of pH
conditions to meet regulatory requirements.
The removal of Zn(II), Pb(II), Cu(II) and Cr(III) was enhanced
between pHs 7 and 11 and a removal efficiency up to 99.37–
99.69% was achieved by the fly ash–lime-carbonation
2613
treatment. The residual concentrations of heavy metals were
(mg L1): Cr 0.08–0.57, Cu 0.14–0.35, Pb 0.03–0.52 and Zn 0.45–
2.11. The heavy metal solutions treated with lime, fly ash and
carbon dioxide met the stringent discharge level of below
0.5 mg L1 for Cu, 2.0 mg L1 for Zn, 1.0 mg L1 for Pb and
1.5 mg L1 for Cr.
The investigation of the sludge produced showed that
calcium–heavy metal double hydroxides including Ca2Cr
(OH)7$3H2O, 2Ca(OH)24Cu(OH)2$H2O and CaZn2(OH)6$2H2O
were formed. In combination with fly ash, lime and CO2, the
sludge volume was reduced and settling performance was
improved significantly. Moreover, the release of heavy metals
from the sludge was decreased greatly by cementation by the
pozzolanic reaction between fly ash and residual calcium
hydroxide and by precipitation of carbonates.
The formation of solid carbonates directly from aqueous
solution containing di- or tri-valent cations and carbon dioxide
(CO2) is a complex process, but may have potential to be
developed for industrial application. The speciation of metals
and their interactions with precipitant ions in solution is an
important area for further study, as it will provide key infor-
mation for optimization of the treatment-step described here.
Acknowledgements
This work was supported by Shanghai Leading Academic
Discipline Project (No: B604) and Natural Science Foundation of
China (No. 50874031). We would like to express our sincere
thanks to the anonymous referees for their constructive
suggestions.
references
Alinnor, I.J., 2007. Adsorption of heavy metal ions from aqueous
solution by fly ash. Fuel 86, 853–857.
Alvarez-Ayuso, E., Garcıa-Sanchez, A., Querol, X., 2003.
Purification of metal electroplating wastewaters using
zeolites. Water Res. 37, 4855–4862.
Baltpurvins, K.A., Burns, R.C., Lawrance, G.A., 1996. Use of the
solubility domain approach for the modeling of the hydroxide
precipitation of heavy metals from wastewater. Environ. Sci.
Technol. 30, 1493–1499.
Charerntanyarak, L., 1999. Heavy metals removal by chemical
coagulation and precipitation. Water Sci. Technol. 39, 135–138.
Chen, Q.Y., Hills, C.D., Tyrer, M., Slipper, I., Shen, H.G., Brough, A.,
2007. Characterisation of products of tricalcium silicate
hydration in the presence of heavy metals. J. Hazard. Mater.
147, 817–825.
Chen, Q.Y., Hills, C.D., Yuan, M.H., Liu, H.H., Tyrer, M., 2008.
Characterization of carbonated tricalcium silicate and its
sorption capacity for heavy metals: a micron-scale composite
adsorbent of active silicate gel and calcite. J. Hazard. Mater. 153,
775–783.
Chen, Q.Y., Tyrer, M., Hills, C.D., Yang, X.M., Carey, P., 2009.
Immobilisation of heavy metal in cement-based solidification/
stabilization: a review. Waste Manage. 29, 390–403.
Cho, H., Oh, D., Kim, K., 2005. A study on removal characteristics
of heavy metals from aqueous solution by fly ash. J. Hazard.
Mater. B127, 187–195.
2614 water research 43 (2009) 2605–2614
Crittenden, J.C., Trussell, R.R., Hand, D.W., 2005. Water
Treatment: Principles and Design. John Wiley & Sons, Inc.,
New Jersey.
Dabrowski, A., Hubicki, Z., Podkoscielny, P., Robens, E., 2004.
Selective removal of the heavy metals from waters and
industrial wastewaters by ion-exchange method.
Chemosphere 56, 91–106.
Doyle, F.M., Liu, Z.D., 2003. The effect of triethylenetetraamine
(trien) on the ion flotation of Cu2þ and Ni2þ. J. Colloid Interface
Sci. 258, 396–403.
Duan, J., Gregory, J., 2003. Coagulation by hydrolysing metal salts.
Adv. Colloid Interface Sci. 100, 475–502.
Environmental Protection Agency, USA, 2002. National
recommended water quality criteria – 2002. US Environment
Protection Agency EPA/822-R-02–047.
Environmental Protection Agency, USA, 2004. National pollutant
discharge elimination system (NPDES) [http://www.cfpub.epa.
gov/npdes/home.cfm?program_id ¼ 3>].
Giles, D.E., Ritchie, I.M., Xu, B.A., 1993. The kinetics of dissolution
of slaked lime. Hydrometallurgy 32, 119–128.
Glasson, D.R., 1961. Reactivity of lime and related oxides: crystal
changes in hydrated lime at different temperature. J. Appl.
Chem. 11, 24–27.
Garcia-Sanchez, A., Alvarez-Ayuso, E., 2002. Sorption of Zn, Cd
and Cr on calcite. Application to purification of industrial
wastewaters. Miner. Eng. 15, 539–547.
Hills, C.D., Sweeney, R.E.H., Buenfeld, N.R., 1999. Microstructural
study of carbonated cement-solidified synthetic heavy metal
waste. Waste Manage. 9, 325–331.
Holt, C.C., Freer-Hewish, R.J., 1996. Lime treatment of capping
layers in accordance with the current specification for
highway works. In: Rogers, C.D.F. (Ed.), Lime Stabilisation.
Thomas Telford, London, pp. 51–61.
Juang, R.S., Shiau, R.C., 2000. Metal removal from aqueous
solutions using chitosan-enhanced membrane filtration.
J. Membr. Sci. 165, 159–167.
Khedr, M.G., 2008. Membrane methods in tailoring simpler, more
efficient, and cost effective wastewater treatment
alternatives. Desalination 222, 135–145.
Kurniawan, T.A., Chan, G.Y.S., Lo, W.H., Babel, S., 2006. Physico-
chemical treatment techniques for wastewater laden with
heavy metals. Chem. Eng. J. 118, 83–98.
Lazaridis, N.K., Matis, K.A., Webb, M., 2001. Flotation of metal-
loaded clay anion exchangers. Part I: the case of chromate.
Chemosphere 42, 373–378.
Lee, D.J., Waite, T.D., Swarbrick, G., 2005. Effect of calcite on lead-
rich cementitious solid waste forms. Cem. Concr. Res. 35,
1027–1037.
Li, Y.J., Zeng, X.P., Liu, Y.F., 2003. Study on the treatment of
copper-electroplating wastewater by chemical trapping and
flocculation. Sep. Purif. Technol. 31, 91–95.
Licsko, I., 1997. Realistic coagulation mechanisms in the use of
aluminium and iron(III) salts. Water Sci. Technol. 36, 103–110.
Lim, S., Jeon, W., Lee, J., Lee, K., Kim, N., 2002. Engineering
properties of water/wastewater-treatment sludge modified by
hydrated lime, fly ash and loess. Water Res. 36, 4177–4184.
Marani, D., Macchi, G., Pagano, M., 1995. Lead precipitation in the
presence of sulphate and carbonate: testing of
thermodynamic predictions. Water Res. 29, 1085–1092.
Matis, K.A., Zouboulis, A.I., Gallios, G.P., 2004. Application of
flotation for the separation of metal-loaded zeolite.
Chemosphere 55, 65–72.
Montes-Hernandez, G., Concha-Lozano, N., Renard, F.,
Quirico, E., 2008. Removal of oxyanions from synthetic
wastewater via carbonation process of calcium hydroxide:
applied and fundamental aspects. J. Hazard. Mater., doi:10.
1016/j.hazmat.2008.11.120.
Montes-Hernandez, G., Perez-Lopez, R., Renard, F., Nieto, J.M.,
Charlet, L., 2009. Sequestration of CO2 by using carbonation
of coal combustion fly-ash. J. Hazard. Mater. 161,
1347–1354.
Ngah, W.W.S., Hanafiah, M.A.K.M., 2008. Removal of heavy metal
ions from wastewater by chemically modified plant wastes as
adsorbents: a review. Bioresour. Technol. 99, 3935–3948.
Ritchie, I.M., Xu, B.A., 1990. The kinetics of lime slaking.
Hydrometallurgy 32, 119–128.
Stumm, W., Morgan, J., 1995. Aquatic Chemistry, third ed. John
Wiley & Sons, Inc., New Jersey.
Sudilovskiy, P.S., Kagramanov, G.G., Kolesnikov, V.A., 2008. Use of
RO and NF for treatment of copper containing wastewaters in
combination with flotation. Desalination 221, 192–201.
Tadesse, I., Isoaho, S.A., Green, F.B., Puhakka, J.A., 2006. Lime
enhanced chromium removal in advanced integrated
wastewater pond system. Bioresour. Technol. 97, 529–534.
Tuenay, O., 2003. Developments in the application of chemical
technologies to wastewater treatment. Water Sci. Technol. 48,
43–52.
Tuenay, O., Kabdasli, N.I., 1994. Hydroxide precipitation of
complexed metals. Water Res. 28, 2117–2124.
Viguri, J., Andres, A., Ibanez, R., Ruiz Puente, C., Irabien, A., 2000.
Characterization of metal finishing sludges: influence of the
pH. J. Hazard. Mater. A79, 63–75.
Wang, S., Wu, H., 2006. Environmental-benign utilisation of
fly ash as low-cost adsorbents. J. Hazard. Mater. B136,
482–501.
Wingenfelder, U., Hansen, C., Furrer, G., Schulin, R., 2005.
Removal of heavy metals from mine water by natural zeolites.
Environ. Sci. Technol. 39, 4606–4613.
Xu, B.A., Giles, D.E., Rithie, I.M., 1997. Reaction of lime with
aluminate-containing solutions. Hydrometallurgy 44,
231–244.
Xu, B.A., Giles, D.E., Ritchie, I.M., 1998. Reactions of lime with
carbonate-containing solutions. Hydrometallurgy 48, 205–224.
Xue, Y.J., Hou, H.B., Zhu, S.J., 2009. Competitive adsorption of
copper(II), cadmium(II), lead(II)and zinc(II) onto basic oxygen
furnace slag. J. Hazard. Mater. 162, 391–401.
Zamboulis, D., Pataroudi, S.I., Zouboulis, A.I., Matis, K.A., 2004.
The application of sorptive flotation for the removal of metal
ions. Desalination 162, 159–168.
Zouboulis, A.I., Kydros, K.A., Matis, K.A., 1995. Removal of
hexavalent chromium anions from solutions by pyrite. Water
Res. 229, 1755–1760.