Desalination 206 (2007) 185–197

Inhibition of scaling in the presence of copper and zinc by

various chemical processes

S. Ghizellaouia*, M. Euvrardb, J. Ledionc, A. Chibania

a
Département de Chimie, Université de Mentouri de Constantine, Route de Ain El Bey , 25000 Constantine, Algérie
Tel./Fax +213 (31) 81 88 16; email: gsamira@yahoo.com
b
LCMI (Laboratoire de Chimie des Matériaux et Interfaces), UFR Sciences et Techniques,
F-25030 Besançon Cedex, France
c
Ecole Nationale Supérieure d’Arts et Métiers, 151, Boulevard de l’Hôpital, 75013 Paris, France

Received 2 February 2006; accepted 16 February 2006

Abstract
The drinking water supply of the town of Constantine is done starting from the subsoil waters of Hamma which
are furring and which have a hardness of 60 French degrees. So the phenomena of scaling are frequent in the
distribution networks of drinking water and in the industrial facilities. They are characterized by adherent crystalline
deposits primarily made up of calcium carbonate. These deposits result from the rise in the temperature of water,
the degasification pulled by certain industrial processes or in the interactions between water and the materials with
which it is in contact. The goal of this work is to study if metal cations Zn2 + and Cu2+ could have impacts on the
formation of the tartar. For that, one uses various experimental protocols which make it possible to define the
furring capacity of water and which were used on a laboratory-scale: tests on polyethylene, tests on scaling potential
and chemical tests. For the tests carried out on polyethylene the total efficiency of 100% was reached for copper
concentration of 1.5 mg/L. On the other hand, for zinc, it was reached at 1 mg/L. The test on scaling potential
developed by Ferreux et al. [1] was implemented after doping water at 1 mg/L for Cu and 5 mg/L for Zn. The rate
of precipitation of CaCO3 for reference water was 25.4%. It was higher than the rate of precipitation of CaCO3 for
water doped with Zn (13.3%) and Cu (20%). Moreover, the analysis of images of the precipitates obtained for
reference and treated water revealed that the formed crystals, especially by doping with Zn, had smaller sizes (7.8
µm) and they were much more important in the solution (423×108) compared to raw water (127.108). This shows
the inhibiting effect of Zn.Taking into account the weak Cu concentration, no significant effect could be evidenced

*Corresponding author.

Presented at EuroMed 2006 conference on Desalination Strategies in South Mediterranean Countries: Cooperation
between Mediterranean Countries of Europe and the Southern Rim of the Mediterranean. Sponsored by the
European Desalination Society and the University of Montpellier II, Montpellier, France, 21–25 May 2006.

0011-9164/07/$– See front matter © 2007 Published by Elsevier B.V.

doi:10.1016/j.desal.2006.02.066
186 S. Ghizellaoui et al. / Desalination 206 (2007) 185–197

during the tests on scaling potential.These conclusions are in agreement with those obtained previously by Ghizellaoui
et al. [2] during chemical tests. Indeed, the results showed that Cu and Zn are very effective inhibitors for Zn
concentration of 5 mg/L and Cu concentration of 3 mg/L with Langelier saturation index (LSI) close to 0.5 whereas
for the reference water it was close to 0.8.Lastly, the analysis under the electron microscope with sweeping (MEB)
and by microsounder showed that the dominant crystalline variety was calcite and copper and zinc distribution was
not homogeneous on calcium carbonate.
Keywords: Drinking water; Scaling; Calcium carbonate; Antiscaling tests; Metal ions

1. Introduction which occur between water and soil and water

and atmosphere. As the geological nature of the
Scaling, frequent phenomenon in distribution
ground conditions affects the chemical com-
of drinking water, is characterized by the appear-
position of natural water, this results in extremely
ance on a surface, metal or not, of an adherent
varied qualities water. Various ions are thus found
mineral deposit primarily made up of calcium car-
in great quantities in natural water (bicarbonates,
bonate. These deposits involve serious technical
chlorides, sulphates, calcium, magnesium, sod-
and economic problems such as obstruction of the
ium, potassium).
drains, scaling of water heatings, seizing of the
According to Ladel [17], the presence of
stopping devices (taps, vannes), etc.
certain ions can form more or less soluble salts
To prevent or limit the formation of the tartar,
which will precipitate by possibly leaving sus-
various techniques make it possible to quantify
pended particles. The deposit obtained in general
the furring capacity of water [1,3,4]. Some are of
consists of calcium carbonate, of silica or calcium
electrochemical type such as tests of chrono-
sulphate. On the other hand, the tartar formed from
amperometry [5] which was the subject of various
not very mineral-bearing natural water, is pri-
developments and various improvements [6]
marily made up of calcium carbonate which is
Moreover, that one is at the base of other methods
the least soluble compound. According to Hort
such as chronoelectrogravimetry [7–9] and the
[18], the consequences of these deposits are of
impedancemetry [10–13].
three types:
The town of Constantine is fed from water
• Hydraulic — by reduction of the by pass sec-
resource of Hamma which has hardness of about
tions of the tubes which involves an increase
60 French degrees and a significant capacity fur-
in pressure losses.
ring. This study is intended initially for the evalu-
• Thermal — the tartar with a bad coefficient of
ation, with the inhibition of the furring capacity,
thermal transfer decreases the quality of the
of this water in the presence of Cu and Zn by using
exchange.
tests on polyethylene, tests on scaling potential
• Mechanical — in comparison with the de-
and the chemical tests. Some researchers have
posited quantities in certain structures, those
already studied the modification of CaCO3 growth
see their mechanical resistance seriously called
in the presence of foreign ions [14–16]. However
in question again.
the study was never made on such hard water.

2. Scaling and precipitation of CaCO3 3. Phases of calcium carbonate precipitation

The mineral compounds present in natural According to Girou [19], Leroy [20] and Leroy
water are essentially originatd from the exchange et al. [21] the precipitation of calcium carbonate
 S. Ghizellaoui et al. / Desalination 206 (2007) 185–197 187

generally comprises two stages: germination and

growth.

3.1.Germination
Germination consists in the appearance of a
solid phase within a solution which is initially de-
prived of it. The appearance of the germs is done
according to the chemical reaction:
Ca 2+ + CO32− → CaCO3 (1)
When germination occurs in water and without Fig. 1. Adsorption of calcium carbonate germs on a metal
wall.
contact with another surface, germination is known
as homogeneous. On the other hand, in contact
with a compound other than CaCO3, which con-
stitutes a solid surface, germination is known as 3.2. Growth of the deposits on the walls
heterogeneous. For the interaction between the In the case of the metal walls, as in the case of
germs of CaCO3 and the walls, Lédion et al. [22] the insulating walls containing polymer, the first
note that when water is sufficiently calcifying adsorbed germs can be used as site of growth for
(supersaturation coefficient δ > 1) and free from excesses of CaCO3 suitable for settling. But after
a significant quantity of impurities, colloidal cal- covering the total surface, there is no more dis-
cium carbonate germs appear. They have a posi- tinction to make between metal wall and insulating
tive potential ζ if such a germ is transferred in the wall. Thus we are in the presence of a calcium
vicinity of a metal wall. Therefore it can then be carbonate wall which turns, in its turn, to behave
adsorbed on the surface, which has, in general, a like an insulating wall. It will be able to be used
negative electrochemical potential. Two cases are as site of growth or to continue to trap by the elec-
then possible: trostatic effect the CaCO3 germs. The growth of
• the germ strongly adheres to the wall and it is the layer continues with this mechanism.
the starter of scaling of the metal wall.
• the germ (of a larger size than in the preceding
case), which is transferred in the vicinity of
the metal wall, is simply attracted without be- 4. Inhibition of CaCO3 formation
ing really adsorbed. In this case, it can be used 4.1. Inhibition of the germination phase
as site of growth of calcium carbonate, locally
According to Ladel [17], at the top saturation
in excess (Fig. 1).
rate of about 40, homogeneous germination cor-
On the other hand, if one is in the presence of responds to energy stabilization of the system and
insulating surfaces, which are not metallic, the cannot thus be inhibited. But the majority of na-
phenomenon remains similar even if the nature tural waters are metastable, their supersaturation
of surfaces involves a different type of adhesion. generally lying between 1 and 40. In this case,
In the case of polymers, only attraction of the precipitation, by mechanism of a radical type, can
electrostatic type can be considered, because the be inhibited by certain organic compounds present
reactivity of the majority of polymer surfaces is in natural water. The inhibitors of germination do
probably null with regard to the local modification not intervene with the growth once of the germs
of calcocarbonics balances. are produced.
188 S. Ghizellaoui et al. / Desalination 206 (2007) 185–197

4.2. Inhibition of the growth following the drainage duct in the distribution sys-
tems. They prevent the nucleation and growth of
According to Lin [23] and Leroy et al. [24]
calcium carbonate crystals.
the best inhibitions were obtained for compounds
such as polyphosphates and some compounds as
organic phosphorus or polymeric acrylic resins. 5.2. Softening in the presence of Zn
Indeed, they present a great affinity for solid
Meyer [14] studied the effect of 34 additives
CaCO3 on the surface of which they are adsorbed
on the growth speed of calcite, among which is
by blocking active sites of germs in growth. Thus,
Zn2+ ion, which proves that this element is a good
this could involve a modification of the electric
inhibitor, more efficient than Cu2+ ion. According
charge which, being initially positive, becomes
to Meyer [14] there is a reduction of the growth
negative. This leads then to a local reduction in
speed by 20% for Zn2+ concentration of 5×10–8 M,
the saturation rate of the medium which becomes
by 50% for 10–7 M and by 80% for 2×10–7 M.
lower than 1. The germination of CaCO3 is always
Urion and Lejeune [28] also noticed the disap-
possible but, as they appear, the germs are inhi-
pearance of scaling on cooling installations in the
bited and cannot grow any more.
presence of Zn. On the other hand, Glasner and
Weiss [29] showed that cations co-precipitate with
5. Chemical softening in the presence of metals calcite. They also supposed the formation of a
complex carbonate ion [Zn(CO3)2]2– which would
Kavanagh et al. [25] and Pernot [16] note the
be used as centre of nucleation for the crystalliza-
effects of the metal ions which affect the growth
tion of calcite.
of calcium carbonate by reducing the speed of nuc-
The inhibiting effect of zinc is significant ac-
leation and growth of the crystal, by deteriorating
cording to the study of Abouali et al. [27]: there
the morphology of the precipitate and, in certain
is inhibition of the germination of calcium carbo-
cases, by causing the precipitation of a less stable
nate starting from 0.02 mg/L of Zn2+.
phase (aragonite or vaterite) in place of calcite.
This assumption was confirmed by Wada et al.
[26]. According to Gutjahr et al. [15], the metal 6. Test of scaling on polyethylene
ions (Fe2+, Zn2+) could be adsorbed on the crystals,
According to Mathieu and Thomas [30], the
thus blocking the sites of growth.
plastics as insulators are furred by an electrostatic
phenomenon. So that an electrostatic attraction
5.1. Softening in the presence of Cu
phenomenon would exist, it is necessary that the
According to the work on the influence of cop- wall and the germ in suspension are of opposite
per on the furring capacity of water, the inhibiting charges. However, natural water has a permittivity
effect of this last is observed for a content which relating to 20°C close to 80 and thus quite higher
is lower than the acceptable maximum value than that of the plastics (relative permittivity
(2 mg/L according to the European directive 98/83). ranging between 2 and 5). Thus, the consequence
Abouali et al. [27] could show, by using the fast is that any polymer in contact with moving water
test of scaling on a polyethylene support, that the will tend to charge negatively. Moreover, accord-
germination of CaCO3 is strongly inhibited from ing to Legrand and Leroy [31], any particle of
200 µg/L of Cu2+. It is practically blocked for a CaCO3 has a positive charge in calcifying water.
content higher than 600 µg/L, and this for a water Therefore, the scaling of plastics is done by elec-
which has a hardness of 27 French degrees. Lédion trostatic trapping of CaCO3 germs which, once
et al. [22] showed the release of the ions of Cu2+, adsorbed on the wall, become charged negatively
 S. Ghizellaoui et al. / Desalination 206 (2007) 185–197 189

due to the frictions of water and also because of bearing and has a rather significant hardness. It is
the relative permittivity of solid calcium carbonate primarily bicarbonated calcic.
which, ranging between 8 and 8.5, remains lower
than that of water. caling is thus not stopped by 7.2. Tests of scaling on polyethylene
the first boiler fur. In this case, Lédion et al. [32]
7.2.1. Material used
also specify that the scaling mechanism is not the
same on metallic surfaces. It could be started by The experimental device is represented in Fig. 2.
an in situ germination in the immediate vicinity The sample, according to Abouali et al. [27], is a
of the surface likely to provide electrons to the polyethylene tube (Φ = 75 mm, height = 50 mm,
cathodic reaction of corrosion which leads to local thickness = 1 mm, mass = 12 g). It is plunged in a
alkalisation and precipitation of calcium carbo- volume of 700 cm3 water and is contained in an
nate. On the contrary, on the insulating surface, it austenitic stainless steel beaker. The agitation is
is necessary that the carbonate has already been ensured by magnetic rod turning at 600 rpm.
precipitated in the colloidal form so that it can
settle by an electrostatic process. 7.2.2. Procedure
Before any handling, the samples were cleaned
7. Material and methods with diluted hydrochloric acid at 50%, then rinsed
with tap water and with demineralized water. They
7.1. Physicochemical characteristics of water of were dried in an oven at 50°C during 20 min.They
Hamma were let to stabilize during 40 min before being
The physicochemical results of the analysis of weighed. The weighing was carried out on a bal-
Hamma water are listed in Table 1. By looking at ance to the tenth of mg. At the same time, a control
these results, we see that water of Hamma has a sample was weighed at each series of measure-
significant relative content of organic matter (TOC ments in order to carry out corrections of weighing
3.84–7.5 mg/L).In addition, it is strongly mineral- on all samples. That made it possible to take into
account the hygroscopy variations.
Table 1 The tests on the water of Hamma were accom-
Hamma water analysis plished in series from 4 to 5 samples per type of
water where, in each test, the reference water (raw
Temperature, °C 31–33 water) was compared with the water treated with
pH 6.95–7.94 copper or zinc at ambient temperature (20°C)
EC, mS/cm 0.9–1.30 during 15 h at 50°C during 2 h 30'. The number
TOC, mg/L 3.84–7.5
of samples can be increased to have a good pre-
CO2, mg/L 16.21–37.05
HCO3–, mg/L 320–432 cision. The process of scaling consists of several
Cl–, mg/L 128–150 stages such as:
SO2–
4 , mg/L 120–212 • The first one consists in filling stainless steel
PO3–
4 , mg/L 0.006–0.02 beakers with 700 ml of raw water and water
NO3–, mg/L 3.90–7 treated with copper or zinc and then placing
NO2–, mg/L
them on magnetic stirrers regulated previously
NH4+, mg/L
TH, mg/L CaCO3 475–600 with precision at 600 rpm using a stroboscope.
Ca2+, mg/L 124–131 • The second one consists in placing the previ-
Mg2+, mg/L 30.48–43.2 ously weighed samples suitably on their lids,
Na+, mg/L 84–115 which are then introduced into water and left
K+, mg/L 2.22–13 there under agitation for 15 h.
190 S. Ghizellaoui et al. / Desalination 206 (2007) 185–197

Fig. 2. Experimental device for the tests of scaling on polyethylene.

• In the third and last step the agitation is stopped of hygroscopy of the room of the balances and
and the lids are removed, after the flow of time the mass of tartar deposited on a sample. It follows
envisaged. Using a grip, the samples are r- the relation:
ecovered and dried in an oven at 50°C during
20 min. They are let to stabilize in the room of M ′ = (mass sample) final
the balances during 40 min. Then the weighing (3)
− (mass sample) initial − D
of the samples, as well as the reference, is
carried out. The expression of the efficiency is then given
by the following relation:
7.2.3.Evaluation of the efficiency
Once the weightings have been carried out, the mb − mt
mass of CaCO3 deposited is determined for each E% = × 100 (4)
mb
sample by taking into account the variation of the
reference. After that, the average for raw water with mb is deposited mass for raw water; mt —
and treated water is calculated. For the blank one deposited mass for treated water.
applies the relation: The treatment has an antiscaling effect for a
positive efficiency, if not the effect is accelerating.
D = (mass reference) final
(2) 7.3.Tests on scaling potential
− (mass reference) initial
D is the quantity taken into account of the variation This technique developed in the Water Chem-
 S. Ghizellaoui et al. / Desalination 206 (2007) 185–197 191

istry Laboratory of of Besançon is based on the

M%
evaluation of the number of crystals present in N= (6)
supersaturated water. According to Pernot et al. 4 / 3πr 3 dQa
[33], the test consists in concentrating the sample
at twice in a rotary evaporator under a pressure of where N is thenumber of crystals present per mole
20 mbar at 20°C. Then water is collected in an of Ca2+; M — precipitated mass (g); Qa — moles
Erlenmeyer glass and let to rest during 24 h in a of Ca2+ contained in the sample before concentra-
drying oven (20°C) (Fig. 3). Water is filtered and tion; r — average radius of the particles (cm); d
the precipitate is dried. — density of the crystalline variety present in the
The crystals are then assembled between blades sample (in our case that of calcite it is equal to
and plates in balsam of Canada. Using an image 2.71 g/cm3).
analyzer (Software Esilab) the morphometric char- At the time of carrying out this test it was
acteristics of these crystals are measured — aver- shown that the more is N number, the weaker is
age diameter and the factor of form and length- furring capacity of water.
ening. The antiscaling effect (E%) can be determined
In parallel, the calcium content is measured in by this test by comparing Nt and Nn of formed
water before concentration and in filtered water. crystals in treated and untreated water.
This enables us to know the rate of precipitation
of calcium carbonate. Nt − N n
E% = × 100 (7)
Nn
[Ca i2 + ] − [Ca 2f + ]
M% = × 100 (5) If E > 0 the effect of the treatment is positive.
[Ca i2+ ]
If E = 0 the effect of the treatment is null.If E < 0
the effect of the treatment is negative.
where [Cai2+] is initial calcium concentration be-
fore passage to the rotary evaporator; [Caf2+] is
7.4. Partial and chemical softening in the pres-
final calcium concentration after filtration.
ence of Cu and Zn
Knowing the mass of CaCO3 precipitated and
the average size of the particles it is then possible With the aim of studying partial and chemical
to determine the number of crystals N present in softening in the presence of Cu or Zn, engines
the sample brought back to a unit quantity of cal- having a capacity of one litre were used. They are
cium present before concentration. made of

Fig. 3. Test of scaling potential.

192 S. Ghizellaoui et al. / Desalination 206 (2007) 185–197

agitation at constant temperature for 72 h. 24 h

after the start of the experiment, the analyses were
carried out by taking a volume of sample using a
syringe to check if the evolution in the course of
time took place, and that by analyzing the chem-
ical parameters (Ca2+, Mg2+, Cl–, SO2– 4
, NO–3, Na+,
K , Cu ). After that, 0.04 g/L lime was introduced
+ 2+

to precipitate temporary hardness, which corres-

ponded to a partial softening of the water of Hamma.
24 h after the lime injection, the second analy-
sis was carried out — the analysis of all the chem-
ical parameters and of Cu to see the effect of metal
considered on the furring capacity of water of
Fig. 4. Engine used in the study of the chemical treat-
Hamma. In the same way, the same experiment
ment. with other metal (Zn) was then conducted in the
area of concentrations quoted previously (Table 2).

• a pipe closed by a septum S;

• R1 grinding where slides a piston in order to 8. Results and discussion
avoid the appearance of the gas phase. After
each sampling, it is inserted by gravity propor- 8.1.Tests of scaling on polyethylene
tionally to the extracted volume. According to Fig. 5, it should be noted that
• R2 grinding receiving electrode pH (Fig. 4). the total efficacy (100%) is reached for a concen-
tration of copper equal to 1500 µg/L, for the tests
The samples were doped with metallic cations
carried out at ambient temperature. On the other
in the field of the concentrations indicated in
hand, the results obtained on zinc indicate that
Table 2.
the total blocking of precipitation is ensured for
At the beginning, water was doped by copper
an addition of 1000 µg/L of Zn (Fig. 6). Therefore,
with various concentrations (0.5, 1, 3 mg/L), then,
an inhibiting effect of Zn is more efficient than
the glass engines were introduced into the thermo-
that of Cu. This result was also confirmed by
regulated bath at 25°C. Initially, one the total car-
Abouali et al. [27].
bon and organics were measured under some care
which consisted in agitating the bottles from out-
8.2.Tests of scaling potential
side with a magnetic rod in order to help mounting
bubbles of air and to eliminate them thereafter, to The results of water doped with the copper and
avoid any error to the measure of the total carbon. zinc and also undoped which is subjected to the
After that, the bottles were maintained under test of scaling potential are given in Table 3.

Table 2
Field of the concentrations considered for metallic cations

Doping element Product used Concentrations used (mg/L) Normalizes (mg/L)

2+
Cu CuSO4, 5H2O 0.5, 1, 3 2
Zn2+ ZnSO4, 7H2O 0.5, 1, 5 5
 S. Ghizellaoui et al. / Desalination 206 (2007) 185–197 193

110

100

90
Efficiency (%)

80

70

60

50
0,6 0,8 1,0 1,2 1,4 1,6 1,8 2,0 2,2

Concentration of Cu (mg/L)
Fig. 5. Influence of Cu2+ addition on polyethylene scaling.

120

100
Efficiency (%)

80

60

40

20

0,0 0,2 0,4 0,6 0,8 1,0 1,2 1,4 1,6 1,8

Concentration of Zn (mg/L)

Fig. 6. Influence of Zn2+addition on polyethylene scaling.

• The mass of CaCO3 precipitated is most sig- • The crystals obtained by doping with zinc are
nificant for the reference water of Hamma of smaller size (7.8 µm) and that was given by
(25.4%) compared to the water doped with Cu the image analysis of the precipitates. More-
(20%) and Zn (13.3%). over, they are much more numerous (423×108)
194 S. Ghizellaoui et al. / Desalination 206 (2007) 185–197

Table 3
Results obtained for reference water and the water doped with Cu and Zn

Parameters Reference water of Hamma Water doped with 1 mg/L of Cu Water doped with 5 mg/L of Zn
Size, µm 11.76 11.92 7.85
% precipitated 25.4 20 13.3
Number, 10–8 127 114 423

compared with raw water (127×108) and the We observed that in the presence of Cu 2+
water doped with copper (114×108) (Figs. 7–9). (Table 4) the inhibition of CaCO3 precipitation
Therefore, the inhibiting effect of Zn is more
pronounced compared to Cu where no antiscaling
effect was noted for the concentration used.
The results reported are the average of four
experiments.

8.3. Partial and chemical softening in the presence

of Cu and Zn
During the tests carried out, we noted a certain
stability of the chemical parameters (Cl–, SO2– 4
,
NO3, Na , K ) at the time of the whole experi-
– + +

ments. We considered it useful to represent the

rates of abatement of the parameters where we
noticed a certain difference between the starting
concentration and the final concentration (Tables 4 Fig. 8. Crystals obtained from the reference water of
and 5). Hamma (observation under the optical microscope).

Fig. 7. Crystals obtained from doped water with 5 mg/L Fig. 9. Crystals obtained from doped water with 1 mg/L
of Zn (observation under the optical microscope). of Cu (observation under the optical microscope).
 S. Ghizellaoui et al. / Desalination 206 (2007) 185–197 195

Table 4
Percentage of reduction of the temporary hardness with lime added and in the presence of Cu2+ ions

Raw water of Hamma Water doped with Water doped with Water doped with
0.5 mg/L Cu2+ 1 mg/L Cu2+ 3 mg/L Cu2+
HCO3–, % 15.87 18 32.53 13.82
Ca2+, % 28.19 27.08 27.64 21.70
Mg2+, % 9.53 1.08 3.41 1.78
Cu2+ — 89.38 94.58 81.10
pH 7.66 7.65 7.64 7.80
LSI 0.511 0.495 0.464 0.673

Table 5
Percentage of reduction of the temporary hardness with lime added and in the presence of Zn2+ ions

Raw water of Hamma Water doped with Water doped with 1 mg/L Water doped with 5 mg/L
0.5 mg/L Zn2+ Zn2+ Zn2+
HCO–3, % 24.16 28.84 31.95 18.15
Ca2+, % 25.48 25.24 26.17 12.97
Mg2+, % 0.94 0.37 0 0
Zn2+ 40.98 94.16 95.35 81.75
pH 7.75 7.73 7.74 7.75
LSI 0.554 0.511 0.493 0.617

occurs only at the highest concentration. 28.2% water obtained is very little furring because it is
of Ca2+ have been precipitated in the presence of partially softened.
lime and 21.7% in the presence of 3 mg/L Cu2+.
The inhibition of Mg2+ precipitation was signifi-
8.4. Analysis with the MEB of the precipitates
cant for all the used concentrations of Cu2+. Only
1–3% of Mg was precipitated in the presence of The analysis with the MEB, under several en-
Cu2+ compared with 9.5% in the absence of Cu2+. largements (×3000, ×10000, ×20000), of the ob-
As in the case of Cu2+, doping with Zn2+ inhi- tained precipitates from raw water (Hamma) and
bited the precipitation of Ca2+ only at one high of the water doped with copper and zinc are given
concentration (Table 5). In this case, the level of in Figs. 10–12.
Ca2+ precipitation was 25–26% in the absence and With raw water, the deposit of CaCO3 obtained
in the presence of 0.5 and 1 mg/L of Zn2+. There- (Fig. 10) presents facies which indicates that it is
fore, doping with 5 mg/L of Zn2+ has led to half about calcite with a traditional crystallization .On
reduction (i.e. to 13% precipitation of Ca2+). the other hand, crystallization is different for the
The results indicated in Tables 4 and 5 show doped precipitates compared to the reference. The
that the computed value of the LSI for raw water distribution of copper and zinc cations is not ho-
and doped with various metals (Cu, Zn) is close mogeneous in calcium carbonate (Figs. 11 and 12).
to 0.5. It is advisable to mention, on the one hand, The inhibiting effect of Cu and Zn results in
that all of these experiments were realized in the the formation of mixed carbonates of calcium and
presence of lime and that, on the other hand, the copper CuxCa1-xCO3 and also for that of mixed
196 S. Ghizellaoui et al. / Desalination 206 (2007) 185–197

Fig. 10. Calcite deposit obtained from the Hamma water. Fig. 12. Calcite deposit obtained after addition of zinc to
water of Hamma.

calcium carbonate on the walls of the drains con-

veying drinking waters. For that, various tests
were conducted at a laboratory scale.
We initially applied the test of polyethylene
scaling in raw and treated waters. This was con-
firmed by calculation of the efficiency, which
reached 100% for copper concentration of 1.5 mg/L
and for zinc concentration of 1 mg/L at ambient
temperature.
Then other tests of scaling potential based on
the crystallization of CaCO3 in the reference water
of Hamma and doped water at 1 mg/L for Cu and
at 5 mg/L for Zn made it possible to evidence the
Fig. 11. Calcite deposit obtained after addition of copper
inhibiting effect of zinc and the analysis of images.
to water of Hamma. This has revealed a smaller size of the crystals
formed with a more significant number of crystals.
Finally, tests of partial and chemical softening
calcium and Zn carbonates: ZnxCa1-xCO3. This re- were condiucted by doping water of Hamma by
sult is confirmed by the analysis with the X-ray. metallic cations (Cu2+ and Zn2+) and in the pre-
The mechanism of inhibition is based on the ener- sence of lime in order to determine the possible
gy of dehydration of Cu2+ and Zn2+ ions which is modification of the furring capacity of water in
more significant than that of the Ca2+ ions, which the presence of these cations.
permits blocking the growth of the formed germs. The results revealed that, with the studied con-
centrations, the abatement of calcium in the case
of doping by zinc and copper are pronounced at
9. Conclusion
high concentrations (3 mg/L for Cu, 5 mg/L for
Our study consisted in studying chemical pro- Zn).
cesses aiming at preventing the incrustation of Moreover, the value obtained for the Langelier
 S. Ghizellaoui et al. / Desalination 206 (2007) 185–197 197

saturation index (LSI) was higher than 0.5 for raw [13] C. Deslouis, C. Gabrielli, M. Keddam, A. Khalil, R.
water of Hamma, thus conferring it high super- Rosset, B. Tribollet and M. Zidoune, Electrochim.
saturation and, consequently, the formation of tar- Acta, 42(8) (1997) 1219–1233.
[14] H. Meyer, J. Crystal Growth, 66 (1984) 639–646.
tar. In addition, this value decreased by doping
[15] A. Gutjahr, H. Dabringhans and R. Lecmann, J.
with metallic cations (Cu and Zn) in the presence Crystal Growth, 158 (1996) 310–315.
of lime. This shows well the efficiency of the treat- [16] B. Pernot, Thèse de doctorat de l’Université de
ment used. Franche-Comté, Chimie Physique, 1997.
[17] J. Ladel, Thèse de doctorat de l’Université de René
Descartes de Paris V, Hydrologie appliqué, 1996.
[18] C. Hort, Thèse de doctorat de l’INSA de Toulouse,
References
1994.
[1] M. Ferreux, F. Remy and A. Vidonne, Tribune de [19] A. Girou, Thèse de doctorat d’Etat Toulouse, 1970.
l’Eau, 567(7) (1994) 17–21. [20] P. Leroy, Tribune de l’Eau, 567(7) jan.–fév. (1994)
[2] S. Ghizellaoui, S. Taha, G. Dorange, A. Chibani and 37–42.
J. Gabon, Desalination, 171 (2004) 133–138. [21] P. Leroy, J. Ladel and J. Baron, TSM, 1 (1995) 43–
[3] C. Gabrielli, M. Keddam, H. Perrot, A. Khalil, R. 48.
Rosset and M. Zidoune, J. Appl. Electrochem., 26 [22] J. Lédion, Y. Jan, Y. Orain and J. Baron, J.I.E.
(1996) 1125–1132. Poitiers, Septembre (2002) 16.
[4] J. Lédion, B. François and J. Vienne, J. Europ. [23] W. Lin, Thèse de doctorat, Paris VI, 1991.
d’Hydrologie, 28(1) (1997) 15–35. [24] P. Leroy, W. Lin, J. Lédion and A. Khalil, Aqua,
[5] J. Lédion, P. Leroy and J.P. Labbé, TSM, juillet– 42(1) (1993) 23–29.
août (1985) 323–328. [25] A. Kavanagh, T. Rayment and T. Price, J. Chem.
[6] J. Lédion, Tribune de l’Eau, 567(7) jan–fev (1994). Soc. Faraday Trans., 86(6) (1990) 965–972.
[7] A. Khalil, P. Sassiat, C. Colin, Ch. Meignen, C. [26] N. Wada, K. Yanashit and T. Umegaki, J. Crystal
Garnier, C.Gabrielli, M. Keddam and R. Rosset, CR Growth, 148 (1995) 279–304.
Acad. Sci., Paris, 314(2) (1992) 145–149. [27] E. Abouali, O. Jean and J. Lédion, J. Europ.
[8] R. Rosset, C. Colin and M. Zidoune, CR Acad. Sci., d’Hydrologie, 27(1) (1996) 109–126.
Paris, 317(2) (1993) 1033–1038. [28] E. Urion and G. Lejeune, L’Eau, 37(2) (1950) 23–
[9] A. Neville, T. Hodgkiess and A.P. Morizot, J. Appl. 28.
Electrochem., 29 (1999) 455–462. [29] A. Glasner and D. Weiss, J. Inorg. Nucl. Chem.,
[10] A. Khalil, C. Colin, C. Gabrielli, M. Keddam and 42(5) (1980) 655–663.
R. Rosset, CR Acad. Sci., Paris, 316(2) (1993) 19– [30] F. Mathieu and J. Thomas, Mémoire d’Ingénieur,
24. 1994, p. 30.
[11] C. Gabrielli, M. Keddam, G. Maurin, H. Perrot, R. [31] L. Legrand and P. Leroy, La technique de l’eau et
Rosset and M. Zidoune, J. Electroanal. Chem., 412 de l’assainissement, 394 (Oct.) (1979) 19–25.
(1996) 189–193. [32] J. Lédion, Y. Guengnon, C. Ribal, P. Combaz and J.
[12] C. Gabrielli, M. Keddam, A. Khalil, R. Rosset and Verdu, TSM, juillet–août (1993) 355–360.
M. Zidoune, Electrochim. Acta, 42(8) (1997) 1207– [33] B. Pernot, M. Euvrard and P. Leroy, 12emes journées
1218. Informations eau (Poitiers), 35 (1996) 1–11.