ARTICLE IN PRESS

WAT E R R E S E A R C H 40 (2006) 61 – 66

Available at www.sciencedirect.com

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

Electrokinetic dewatering of Turkish glass sand plant

tailings

O. Bayata,, O. Kilica, B. Bayatb, M. Anila, H. Akarsuc, C. Pooled

a
Mining Engineering Department, Cukurova University, Balcali, 01330 Adana, Turkey
b
Environmental Engineering Department, Cukurova University, Balcali, 01330 Adana, Turkey
c
CAMIS Mining Co., Mersin, Turkey
d
Department of Mining and Mineral Engineering, University of Leeds, Leeds LS2 9JT, UK

art i cle info ab st rac t

Article history: In this study, the dewatering of glass sand plant tailings from Mersin, Turkey was
Received 19 November 2003 investigated using an electrokinetic technique. The particle size ðd80 Þ of the solid waste
Received in revised form material tested was less than 0.020 mm and consisted mainly of silica, orthoclase, alumina,
18 July 2005 potassium and iron oxides. In current plant practice, Larox high-pressure filters are used to
Accepted 18 October 2005 produce a filter cake containing 22–25% moisture. As an alternative, a laboratory-scale
dewatering pond using an electrokinetic technique was designed and a final product (cake)
Keywords: containing 34% moisture was obtained after 24 h treatment using 14.6 V direct current
Tailings disposal applied for 40 min. When the treatment continued up to 48 h, a firm-to-hard cake was
Dewatering produced.
Flocculation & 2005 Elsevier Ltd. All rights reserved.
Recycling
Waste processing

1. Introduction
Silica sand occurs in abundance in varying degrees of purity,
but only a small percentage is suitable for glassmaking and
other high-grade industrial applications, including foundry,
ceramics, chemicals, fillers and the construction industry.
Mineral processing is an essential step in upgrading the raw
sand to meet industrial specifications, such as a high silica
content (495.4% SiO2), narrow size distribution and small
amount of clay combined with low levels of TiO2, Al2O3 and
Fe2O3. Sedimentation of glass sand tailings is frequently
difficult and lengthy, owing to their nature and small particle
size. Chemical dosing with metal ions is often used in
primary treatment to remove colloidal materials from waste-
water, but new technologies such as electrocoagulation (EC)
are now able to produce flocs of larger size and density
(Newman, 1984; Koren and Syverson, 1995; Mollah et al., 2001;
Larue and Vorobiev, 2003). It is generally accepted that the EC
process involves three successive stages: (a) formation of
coagulants by electrolytic oxidation of the ‘sacrificial elec-
trode’; (b) destabilization of the contaminants, particulate
suspension, and breaking of emulsions; and (c) aggregation of
the destabilized phases to form flocs. The destabilization
mechanism of the contaminants, particulate suspension, and
breaking of emulsions has been described in broad steps and
may be summarized as follows (Mollah et al., 2001):
(i) Compression of the diffuse double-layer around the
charged species, which is achieved by the interac-
tions of ions generated by dissolution of the sacrificial

Corresponding author. Tel.: +90 322 3386119; fax: +90 322 3386126.
E-mail addresses: obayat@cu.edu.tr (O. Bayat), zenkilic@cu.edu.tr (O. Kilic), bbayat@cu.edu.tr (B. Bayat), manil@cu.edu.tr (M. Anil),
hakarsu@sisecam.com.tr (H. Akarsu), c.poole@leeds.ac.uk (C. Poole).
0043-1354/$ - see front matter & 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.watres.2005.10.021

62 WAT E R R E S E A R C H
electrode, due to passage of current through the
solution.
(ii) Charge neutralization of the ionic species present in
wastewater, which is caused by the counter ions, produced
by the electrochemical dissolution of the sacrificial elec-
trode. These counter ions reduce the electrostatic inter-
particle repulsion sufficiently so that the van der Waals
attraction predominates, thus causing coagulation.
(iii) Floc formation, and the floc formed as a result of
coagulation creates a sludge blanket that entraps and
bridges colloidal particles that have not been complexed.
Differences in the mobilities of ions, molecules, or particles
in an electric field can also be exploited to perform useful
separations. For electrolytic solutions, the migration of
charged species in an electric field constitutes an additional
mechanism of mass transfer. Thus, the flux of an ionic species
Ni in g mol/cm2 s in dilute solutions can be expressed as
Ni ¼
zi ui Ici rE
Di ci þ ci v,
where the ionic mobility ui is the average velocity imparted to
the species under the action of a unit force (per mole); v is the
stream velocity in cm/s. In the present case, the electric field
rE is in V/cm and the charge zi I per mole, where I is the
Faraday constant in C/g equivalent, Di is the ionic-diffusion
coefficient, cm2/s, and zi is the valence of the ith species.
Multiplication of this force by the mobility and the concen-
tration ci (g mol/cm3) yields the contribution of migration to
the flux of the ith species (Robinson and Stokes, 1959). Particle
sedimentation can be accelerated by imposing an electrical
field in the proper orientation; that is, for negatively charged
particles, horizontal placement of the anode (positive elec-
trode) near the bottom of the slurry, with the negatively
charged cathode floating on the surface (Sprute and Kelsh,
1974). When sedimentation is complete, further drainage can
-0.106 mm
Larox Filter
Filtrate
Electrokinetic
Dewatering
Dewatered Tailings (20-22% moisture)
Fig. 1 – Flowsheet of glass sand-processing plant.
ARTICLE IN PRESS
40 (2006) 61– 66
be stimulated by reversing the polarity, with the bottom
cathode acting as a drain. In an electrical field established
between two separated electrodes buried in slurry, the solid
particles will not move appreciably because of their relatively
dense packing, but the water will be carried toward the
negatively charged cathode by the viscous drag of the
migrating positive ions. The effectiveness of electrokinetic
techniques (direct current application) in consolidating slur-
ries of fine-grained solids in watery suspensions has been
investigated since the 1970s (USBM, 1974a, b, 1975; Mollah
et al., 2001, 2004; Larue and Vorobiev, 2003; Larue et al., 2003).
Matteson et al. (1995) investigated the EC of aqueous
suspensions of ultrafine kaoline particles in a single stirred
cell apparatus and in a continuously flowing system consist-
ing of three cells in series. By taking advantage of the
electrical surface charge on solid particles in water, fast
particle settlement and sediment dewatering can be achieved
by imposing a relatively large dc voltage (60–240 V/m) between
electrodes placed in the slurry. This technique has already
(1)
been successful for treating siliceous tailings from Idaho,
USA, metal mines. Slurry properties found to affect response
included electrical conductivity, particle size distribution,
water content, and surface charge density. Chemical proper-
ties also influenced behaviour, but their importance was
specific to each sample and difficult to characterize (Sprute
and Kelsh, 1980).
Camis Madencilik AS has been operated in Mersin (a
southern province of Turkey) since 1970. The ore contains
quartz (88%), orthoclase (10%), iron oxides (0.25%) and trace
elements such as Zr, Cr, Ca and Ti (Bayat and Akarsu, 2002).
As shown in Fig. 1, froth flotation is used for the removal of
the colouring impurities, including pyrite, garnet, chromite,
ilmenite, haematite, mica, etc. The capacity of the concen-
trator is 450,000 tonnes per year and tailings (approximately
50,000 tonnes per year) are thickened and filtered by Larox
Feed (glass sand; -2 cm)
Grinding (Rodmill)
Hydrocyclone
+0.6-0.106 mm
H2SO4
Froth Flotation
Aero 801+825; 370 g/t
(from Cytec Int., Holland)
Tailings Concentrate Superfloc A 110; 70 g/t
Spiral Classifier Alcopol; 80 g/t
Dewatered Tailings Hoppers
(20-25% moisture)
Process
water
 ARTICLE IN PRESS
WAT E R R E S E A R C H 40 (2 006 ) 6 1 – 66 63

high-pressure filters. A filter cake containing 22–25% moisture
is conveyed to the tailings disposal area and the current
dewatering cost is about $1/tonne.
In this study, dewatering of the tailings from the glass sand-
processing plant using an electrokinetic technique was
investigated on a laboratory scale.
2. Materials and methods
Physical and chemical properties of the solid waste (tailings)
before treatment are summarized in Table 1. Chemical
analysis of the sample was carried out at the Camis plant
by XRF spectrometry using a method described in detail by
Skoog et al. (1998). The particle size of the sample ranged
from 0.005 to 0.105 mm and the d80 was less than 0.020 mm. X-
ray measurements were performed using a Philips X-ray
difractometer employing nickel-filtered CuKa radiation. X-ray
diffraction data for crystalline phases of the tailings are given
in Table 2. The mineralogy was found to be mainly orthoclase
(KAlSi3O8), alumina (Al2O3), potassium (K2O), iron oxides
(Fe2O3) and silica (SiO2). The zeta potential (ZP) measurements
of the suspensions were made in distilled water of a constant
ionic strength (0.01 M NaCl) by Zeta-Meter 3.0. Prior to ZP
measurements powder sample was dispersed at 0.1% (w/v)
solid ratio in distilled water, pH and ionic strength were
adjusted to desired values and stirred on a magnetic stirrer
for 5 min at room temperature. Required amount of suspen-
sion was transferred into the electrophoresis cell of the zeta
meter. During the ZP measurement 10 particles were tracked
and the average of their ZP values was taken. Consequently,
the ZP changes of the suspensions as a function of pH were
determined. The working principle of the zeta meter 3.0 is
based on the electrophoresis method and it computes
automatically the ZP using the Smoluchowski equation.
Laboratory electrokinetic dewatering tests were conducted
at the Camis plant using a specially built Plexiglas model (see
Fig. 2) 112 cm long  24 cm wide  24 cm deep. Two series of
tests were performed using various dewatering arrangements
(Kilic, 1998). Optimum results were achieved with the follow-
ing procedure. Two iron pipes, with an outside diameter of
2.1 cm and 36 cm long, placed on the horizontal floor and
embedded in gravel, served as a cathode. A 36 cm  20 cm
rectangular grid of a 0.95 cm iron rod acted as the moveable
anode: this was placed on the surface of the cell. For each test,
a 32 litre-suspension containing 3–5% solids was taken from
the plant thickener and poured into the cell after 10 min of
mixing. In the first stage of a two-stage dewatering procedure,
power was applied to the front section at a rate of 14.5 V
between electrodes, with an average current density of
11.88 A/m2 until drainage ceased. For the second stage (after
10 min), the anode was moved to a position 6 cm below the
surface over the cathode, and power was applied at 14.6 V
between electrodes at an average current density of 5.63 A/m2
until water discharge from the pump well ceased. The power
was then switched off after 40 min and the cake was left for
24 h for additional settlement. The final product in the model
was sampled for moisture content. During electrolysis, the
positive side underwent anodic reactions, while on the
negative side, a cathodic reaction was encountered. Consum-
able metal plates, such as iron or aluminium, were usually
used as sacrificial electrodes to continuously produce ions in
the system. The released ions neutralized the charges of the
particles and thereby initiated coagulation.
3. Results and discussion
3.1. Zeta potential measurements
The surface charge characteristics of precipitated metal
hydroxides are of great importance in coagulation. In
common with oxides and other minerals they show an
isoelectric point (i.e.p.) at which the apparent (electrokinetic)
surface charge is zero. At pH values below the i.e.p. the

Table 1 – Properties of the suspension of the glass sand tailings

Property Values

Conductivity (3
4% solids, pH 8, 2372 1C) 12 mS/cm

Solid content 3–4% by weight
Zeta potential at 2372 1C
3% solids, at pH 4
185 mV
at pH 6
120 mV
at pH 8 35 mV
5% solids, at pH 4
151 mV
at pH 6
55 mV
at pH 8 58 mV
Grain-size distribution 94% finer than 0.075 mm
80% finer than 0.020 mm
50% finer than 0.010 mm
pH at 2372 1C 7.4
(% by weight)
SiO2 Al2O3 Fe2O3 TiO2 CaO MgO Na2O K2O Cr2O3 MnO Zn Ni As LOIa
75.12 7.26 2.86 0.48 2.09 1.06 0.03 3.22 2.02 0.04 0.27 0.12 0.17 4.11

a
LOI, loss on ignition.
 ARTICLE IN PRESS
64 WAT E R R E S E A R C H 40 (2006) 61– 66

Table 2 – XRD data for crystalline phases of the tailings observed at a pH of 7.7 and 8.8, respectively. These values are
in accordance with the values reported by earlier researchers
Crystalline phase d spacing (Å)
(Syed Ahmed, 1975; Rudzinski et al., 1992). Below the pH of
Fe2O3_A 2.7000 i.e.p., precipitates of oxy-hydroxides are positively charged.
FeO(OH)_B 2.6900 From the results it is apparent that the magnitude of positive
Al2O3_A 2.5520 ZP of iron and aluminium oxy-hydroxides reduces in the
FeTiO3_B 2.5440 presence of sulphate and sulphite ions without any shift in
Fe3O4_A 2.5320
i.e.p. This clearly indicates that there is no chemical interac-
FeCr2O4 2.5200
Fe2O3_B 2.5190
tion between these ions and metal hydroxy precipitates. At
ZrSiO4 2.5180 the same time, a decrease in the magnitude of ZP of metal
FeO 2.4900 hydroxides in the presence of these ions clearly suggests an
SiO2_qrz 2.4570 electrical interaction. This reduction in positive charge may
FeO(OH)_A 2.4500 be due to a weak electrical interaction between negatively
FeS2_B 2.4300 charged sulphite and sulphate ions and positively charged
Cryolite_C 2.4340
oxy-hydroxy species of iron and aluminium. Transmission
Cu2O 2.4700
FeTiO3_A 2.7540 electron microscopic (TEM) studies conducted by earlier
FeS2_A 2.7090 researchers (He et al., 1996) have revealed that at a pH of
FeO(OH)_B 2.6900 6.0, the aggregated Fe(OH)2 colloids exist as three-dimen-
Fe2O3_A 2.7000 sional (branched or spread out) porous sponge-like precipi-
Al2O3_A 2.5520 tates. Thus, better dewatering under slightly acidic conditions
FeTiO3_B 2.5440
can be explained in the light of the ZP results.
Fe3O4_A 2.5320
Ca2_A 2.5990
MnO 2.5680
Cr2O4 2.5800 3.2. Electrokinetic dewatering
CaSO4_h2 2.5966
Iron upon oxidation in an electrolytic system produces iron
hydroxide, Fe(OH)n, where n ¼ 2 or 3. When a dc current flows
between two iron electrodes, the following electrolysis
reactions are expected to occur in the vicinity of the
Anode electrodes (Mollah et al., 2001):
(Oxidation) Anode:
+
Colloidal Particle FeðsÞ ! Fe2þ

ðaqÞ þ 2e , (2)
Fe2+
E DC power
Fe(OH)2 supply
OH−
Suspension Fe2þ

ðaqÞ þ 2OHðaqÞ ! FeðOHÞ2ðsÞ . (3)
H2
Cathode
Cathode:
(Reduction)
-
Coagulation+Flocculation Sediments 2H2 OðlÞ þ 2e
! 2OH

ðaqÞ þ H2ðgÞ . (4)

Fig. 2 – Schematic of longitudinal cross section of test cell. Overall:

FeðsÞ þ 2H2 OðlÞ ! FeðOHÞ2ðsÞ þ H2ðgÞ . (5)

The Fe(OH)n(s) formed remains in the aqueous stream as a

precipitate is positively charged and at higher pH values it has
a negative charge (Duan and Gregory, 2003).
The ZPs of the suspension versus pH are given in Table 1.
The isoelectrical point at about pH 7 was found at 5% solids
and the magnitude of the ZP increased dramatically with
increasing pH of the suspension. The sample consists of over
75% SiO2 (see Table 1), and may thus be considered as a quartz
mineral. The ZP of quartz is zero at pH 2.0 (Fuerstenau and Jia,
2004). However, when their surface properties, especially ZP
(Table 1), associated impurities i.e., colouring metal oxides
(Fe2O3, TiO2) or other metal oxides (MgO, Al2O3), and geologic
formation are considered, they may indicate some physical,
chemical and mechanical differences with regard to the pure
quartz mineral.
Murugananthan et al. (2004) have revealed that the i.e.p. of
iron oxide/hydroxide and aluminium oxide/hydroxide was
gelatinous suspension, which can remove the pollutants from
wastewater either by complexation or by electrostatic attrac-
tion, followed by coagulation. In the surface complexation
mode, the pollutant acts as a ligand (L) to chemically bind
hydrous iron:
L-HðaqÞ ðOHÞOFeðsÞ ! L-OFeðsÞ þ H2 OðlÞ . (6)
The prehydrolysis of Fe3+ cations also leads to the forma-
tion of reactive clusters for water treatment. Structural
models for these oxyhydroxy iron cations have been
extensively reported in the literature (Schneider and Schwyn,
1987). For instance, wastewater containing Cr6+ (CrO2– 4 ) ions
can be removed by the EC technique using iron as the
sacrificial anode. The ferrous ion (Fe2+) generated by electro-
oxidation of the iron anode can reduce Cr6+ to Cr3+ under
alkaline conditions and is itself oxidized to ferric (Fe3+) ion
 ARTICLE IN PRESS
WAT E R R E S E A R C H 40 (2 006 ) 6 1 – 66 65

according to and 8, respectively, after 10 h (see Fig. 3) with an applied

current density of 5.63–11.88 A/m2. The tests were performed
CrO2
2þ
4 ðaqÞ þ 3Fe ðaqÞ þ 4H2 OðlÞ with the bottom electrode negatively charged, thereby draw-
! 3Fe3þ ðaqÞ þ Cr3þ ðaqÞ þ 8OH

ðaqÞ ð7Þ ing water downwards. The power consumption (costing $0.13/
kWh) required to obtain these results ranged between 1.37
or
and 2.11 kWh per tonne of final product.
CrO2
2þ
4 ðaqÞ þ 3Fe ðaqÞ þ 4H2 OðlÞ The overall efficiency was poor in these initial experiments,
þ 4OH
  
ðaqÞ 3FeðOHÞ3 þ CrðOHÞ3 . ð8Þ so several modifications were made. Three more by-pass
holes were opened on the right side of the cell, allowing water
The Cr3+
(aq) ion is then precipitated as Cr(OH)3(s) by raising the
2

to be pumped more efficiently. A thick layer of sand, 1.5 cm
pH of the solution. The Fe2+ (aq) ions can also reduce Cr2 O7ðaqÞ
deep, was placed on top of the gravel to provide additional
under acidic conditions according to the following reaction:
filtering. The cycle consisted of applying potential to the
Cr2 O2
2þ upper electrode for 15 min, followed by 5 min of no power.
7 ðaqÞ þ 6Fe ðaqÞ
3þ 3þ This 20-min cycle was repeated throughout the dewatering
þ 14Hþ
ðaqÞ 2Cr ðaqÞ þ 6Fe ðaqÞ þ 7H2 OðlÞ . ð9Þ
period. The effects of pH and solid content of the suspension
The generation of iron hydroxides is followed by an using the modified model are shown in Fig. 4. The solid
electrophoretic concentration of particles containing mainly content of the cake was readily increased from an initial 3–5%
mica, clays, iron oxides and silica particles, which are swept to 50%, 62% and 50% at pH 4, 6 and 8, respectively, after 10 h
by the electric field into the region close to the anode. (see Fig. 4) with an applied current density of 5.63–11.88 A/m2.
Particles interact with the iron hydroxides and can be These tests demonstrated that a final product containing 34%
removed either by surface complexation or electrostatic moisture could be obtained after 24 h treatment at pH of 6 for
attraction. Therefore, the EC process improves coagulation a power expenditure of 2.15 kWh per tonne final product.
(Larue and Vorobiev, 2003). Larue and Vorobiev (2003) have indicated that the flocs
For initial experiments, tailings taken directly from the obtained by this technique are larger or have a higher
plant thickener were poured into the model and after effective density when compared with those generated by
complete settlement of solids, the moisture content of the the addition of chemicals. In a recent article Jiang et al. (2002)
product was determined as 62% after 24 h. The effects of pH reported the treatment of wastewater by a hybrid EC
and solids content of the suspension were also evaluated with technique using aluminium electrodes. The authors designed
the laboratory test model. The slimes responded well to two different EC reactors to evaluate their water treatment
electrokinetic dewatering procedures in these small-scale performance and found out the best configurations to
laboratory tests. The solid content of the cake was readily maximize the treatment efficiency. The EC reactors were
increased from an initial 3–5% to 39%, 46% and 44% at pH 4, 6 configured with water up-flow or cross-flow arrangements.

62
70
Initial 58
68 pH = 4
Moisture of the cake (%)

pH = 6
Moisture of the cake (%)

66 54
pH = 8
64
62 50
60 46
58
42
56
54 38 Initial
pH = 4
52
34 pH = 6
50 pH = 8
180 210 300 450 600 30
Dewatering time (min) 180 210 300 450 600 1200 1440
Dewatering time (min)

Fig. 3 – Effects of pH and suspension solid content on

dewatering of glass sand process tailings at Fig. 4 – Effects of pH and suspension solid content after
2372 1C. modification of the model at 2372 1C.
 ARTICLE IN PRESS
66 WAT E R R E S E A R C H
Kovatcheva and Parlapanski (1999) applied a low-frequency
(25–75 Hz) sonic field at current densities of 0.1–0.4 A/m2 to
improve the performance of the EC cell with carbon steel
electrodes. It has been reported to increase the quantity of
iron hydroxides, and thereby enhance coagulation. Using
optimized conditions, after 24 h treatment, a final cake
containing 34% moisture was obtained, but the product could
be converted into a firm-to-hard cake with 20% moisture
content when the process time was extended to 48 h.
As a result of all of these test modifications, the following
operating procedures were found to be necessary:
 water should be constantly pumped from the sump, since
an accumulation at the cathode prevents efficient dewa-
tering,
 cathodes should be embedded on the floor of the settling
pond,
 cathodes must be well embedded to ensure good electrical
contact throughout the experiment when lowering the
anode onto the fill, and
 natural drainage should be initiated as soon as the
material has thickened sufficiently to discharge clean
water; this reduces electrokinetic treatment time and
results in a better final product.
These encouraging results led to a plant-scale electrokinetic
system being suggested as an alternative to the Larox plant
filters for dewatering the glass sand tailings. An evaluation of
the new system showed that six large ponds (55  30  16 m)
would be required for efficient tailings disposal at a predicted
lower overall disposal cost of $0.89/tonne. One of the most
important considerations in lowering the cost of the EC
process is to minimize the IR-drop to enhance current
efficiency. Evolution of oxygen at the anode represents an
unwanted leakage of current, which reduces the overall
current efficiency. This problem can be minimized through
the choice of an anodic material that presents a high over
potential for the oxygen evolution reaction near the anode.
This high over potential allows the realization of anodic
reactions, which are not possible with other materials and in
particular, the oxidation of organic compounds with a very
high current efficiency.
4. Conclusions
Laboratory-scale dewatering experiments demonstrated that
an electrokinetic process could effectively dewater and
densify unclassified tailings from a glass sand processing
plant with moderate power consumption. A-firm-to-hard
cake with 20% moisture content was obtained when the
process time was extended to 48 h.
Acknowledgements
The authors extend their sincere thanks to Camis Madencilik
AS, Mersin, Turkey, for their technical support. This study was
supported by the Cukurova University Scientific Research
Projects Directorate (project no. FBE.98.YL.65).
40 (2006) 61– 66
r e f e r e n c e s
Bayat, O., Akarsu, H., 2002. Evaluation of new collectors for silica/
glass sand and statistical analysis of plant trials. Min. Eng. 15,
293–296.
Duan, J., Gregory, J., 2003. Coagulation by hydrolysing metal salts.
Adv. Colloid Interf. Sci. 100–102, 475–502.
Fuerstenau, D.W., Jia, R., 2004. The adsorption of alklpyridinium
chlorides and their effect on the interfacial behavior of
quartz. Colloids Surf. A: Physicochem. Eng. Aspects 250,
223–231.
He, Q.H., Leppard, G.G., Paige, C.R., Snodgrass, W.J., 1996. Water
Res. 30 (6), 1345.
Jiang, J.Q., Graham, N., Andre, C., Kelsall, G.H., Brandon, N., 2002.
Laboratory study of electrocoagulation–flotation for water
treatment. Water Res. 36, 4064–4078.
Kilic, O., 1998. Dewatering of glass-sand processing plant tailings
with direct current. M.Sc. Thesis, Cukurova University, Adana,
Turkey, 80pp (in Turkish).
Koren, J.P.F., Syverson, U., 1995. State-of-the-art electrofloccula-
tion. Filtr. Sep. 23 (2), 153–156.
Kovatcheva, V.K., Parlapanski, M.D., 1999. Sono-electrocoagula-
tion of iron hydroxides. Colloid Interf. A: Physicochem. Eng.
Aspects 149, 603–608.
Larue, O., Vorobiev, E., 2003. Floc estimation in iron induced
electrocoagulation and coagulation using sedimentation data.
Int. J. Miner. Process 71 (1–4), 1–15.
Larue, O., Vorobiev, E., Vu, C., Durand, B., 2003. Electrocoagulation
and coagulation by iron of latex particles in aqueous suspen-
sions. Sep. Purif. Technol. 31, 177–192.
Matteson, M.J., Dobson, R.L., Glenn Jr., R.W., Kukunoor, N.S., Waits,
W.H., Clayfield, E.R., 1995. Electrocoagulation and separation of
aqueous suspensions of ultrafine particles. Colloids Surf. A:
Physicochem. Eng. Aspects 104, 101–109.
Mollah, Y.M.A., Schennach, R., Parga, J.R., Cocke, D.L., 2001.
Electrocoagulation (EC)—science and applications. J. Hazard.
Mater. 84 (1), 29–41.
Mollah, Y.M.A., Morkovsky, P., Gomes, J.A.G., Kesmez, M., Parga, J.,
Cocke, D.L., 2004. Fundamentals, present and future perspec-
tive of electrocoagulation. J. Hazard. Mater. 114, 199–210.
Murugananthan, M., Bhaskar Raju, G., Prabhakar, S., 2004.
Removal of sulfide, sulfate and sulfite ions by electro
coagulation. J. Hazard. Mater. 109, 37–44.
Newman, J., 1984. Novel separation processes. In: Perry, R.H.,
Green, D. (Eds.), Perry’s Chemical Engineers’ Handbook.
McGraw-Hill, New York.
Robinson, I.R., Stokes, K., 1959. Electrolyte Solutions. Academic
Publishers, New York.
Rudzinski, W.R., Partyka, S., Thomas, S., Bottero, J.Y., 1992.
Langmuir 8, 1154.
Schneider, W., Schwyn, B., 1987. Aquatic Surface Chemistry.
Wiley/Interscience, New York.
Skoog, D.A., Holler, F.J., Nieman, T.A., 1998. Principles of Instru-
mental Analysis, 5th ed. Harcourt Brace & Company, New
York.
Sprute, R.H., Kelsh, D.J., 1974. Laboratory experiments in electro-
kinetic densification of mill tailings. USBM RI 7892, 72pp.
Sprute, R.H., Kelsh, D.J., 1980. Dewatering fine-particle suspen-
sions with direct current. In: Somasundaran, P. (Ed.), Fine
Particle Processing. SME, New York.
Syed Ahmed, M., 1975. In: Somasundaran, P., Grieves, R.B. (Eds.),
Advances in Interfacial Phenomena of Particulate/Solution/
Gas Systems. Application to Flotation Research, vol. 75.
A.I.Ch.E. Symposium Series, 150. AIChE, New York.
USBM RI 7900, 1974a. 43pp.
USBM RI 8197, 1974b. 68pp.
USBM RI 8034, 1975. 47pp.