Journal de la Société Chimique de Tunisie, 2013, 15, 175-181 175

COAGULATION TREATMENT BY Al2 (SO4)3 AND RESIDUAL Al

DETERMINATION IN MEDJERDA WATER DAM (TUNISIA)
Mouna JAOUADI*, Noureddine AMDOUNI

Université Tunis El Manar UR. Physico-Chimie des Matériaux Solides

Faculté des Sciences de Tunis, Manar II 2092 Tunis, Tunisia.

(Reçu le 15 janvier 2013, accepté le 08 Mars 2013)

ABSTRACT: The effect of pH, agitation speed and coagulant dosage on the coagulation performance of aluminium
sulfate with respect to the treatment of Medjerda water was investigated in this paper. The measurement of residual
aluminium speciation was also conducted. The evaluation of treatment efficiency of the optimization of coagulation
process while was determined by measuring the reduction of turbidity, chemical oxygen demand (COD) and organic
matter (OM). The optimal operational conditions are as follows: initial pH value of 6, concentration of aluminium
sulfate of 30 mg.L-1 and agitation speed of 80 rpm.

Keywords: Raw water, coagulation, Aluminium sulfate, chemical oxygen demand, organic matter.

RESUME : L'effet du pH, de la vitesse d'agitation et de la dose du coagulant sur le rendement de coagulation de l’eau
du canal Medjerda ont été étudiés. La mesure d’aluminium résiduel et sa spéciation a également été menée.
L'évaluation de l'efficacité du traitement de l'optimisation des processus de coagulation a été déterminée en mesurant la
réduction de la turbidité, la demande chimique en oxygène (DCO) et de la matière organique (MO). Les conditions
optimales sont les suivantes: pH optimum est égal à 6, la concentration optimale de sulfate d'aluminium est de l’ordre
de 30 mg.L-1 et la vitesse d'agitation est égale à 80 tours par minute.

Mots clés: Eau brute, coagulation, sulfate d’aluminium, demande chimique en oxygène, matière organique.

INTRODUCTION
Coagulation by metal salts such as aluminium sulfate is an important and widely applied
process in water treatment. It is popular as a pre-treatment process for the removal of raw water
turbidity (Tu), organic matter (OM), color, and microorganism [1-2]. Although effective for
removing turbidity and organic materials, Al-based coagulants, particularly alum, may result in
elevated concentrations of residual Al in finished water [3-4]. High concentrations of residual
dissolved Al indicate incorrect coagulant dosing, inefficient use of the coagulant, or problems in the
chemistry of treatment process. Thus, Al fractionation measurements can provide the treatment
plant operator with valuable information about various aspects of plant performance [3].
Aluminium (Alum) determination at low level is of particular interest in potable water units because
this metallic ion is commonly used as reactant for coagulation–flocculation in the treatment of raw
waters to remove colloidal or suspended particles or to eliminate organic matter. At the outlet of
these units, maximum tolerable level of this cation has been fixed to 200 µg L −1 by European
Legislation [5]. This cation is associated with various health problems in numerous studies, from
gastrointestinal damage and phosphate deficiency to dialysis encephalopathy, renal oestrodistrophy
and Alzheimer’s disease [6-7]. In order to optimize coagulation process in drinking water plants and
to minimise aluminium levels in finished water, monitoring of this metal contents during and after
raw water treatment is there- fore needed. Up to now, the coagulant quantities are generally

*
Corresponding author, e-mail : ing.mouna@gmail.com
176 Mouna Jaouadi et al., J. Soc. Chim. Tunisie, 2013, 15, 175-181

determined by the empiric Jar-test technique that induces problems of excess (or insufficient)
reagent, particularly during period of fast variation in water quality [8-9].
The combination of coagulation with other appropriate physicochemical treatment processes such
as water initial pH, coagulant dosage, and mixing conditions results in enhanced efficiency of the
coagulation–flocculation process [10-11]. During raw water treatment, the determination of
coagulant dosage is one of the most important factors among various works conducted in unit
processes [12]. Other properties of the raw water such as influent turbidity, agitation speed, water
initial pH, etc. affects the type of coagulant and its dosage [13-14].
The aim of this study was to optimize coagulation-flocculation. This study examined the individual
and interactive influences of coagulation and flotation factors (i.e., speed agitation, coagulant dose,
and pH) on the removal of turbidity, COD, and OM. The content and variation of different residual
Al speciation in the water after coagulation with reference to the specific surface water treatment
were systematically examined and analyzed.

2. MATERIALS AND METHODS

2.1. Materials
Analytically pure sodium hydroxide (NaOH) and hydrochloric acid (HCl) of concentration 0.50
molL-1 reagents were used for regulating the pH of domestic sewage.
Al2 (SO4)3.18H2O (Aldrich) has been used as coagulant. The initial Al2 (SO4)3 solution was prepared by
directly dissolving certain amount of Al2 (SO4)3.18H2O to deionized water and it was used immediately after
preparation.

2.2 COD (chemical oxygen demand) measurement

Sample for COD measurement was stored in acid- cleaned and muffled borosilicate clear glass vials
with Teflon-lined caps, and 50 μl of 6 N HCl was added (pH < 2) to remove inorganic carbon and inhibit
bacterial activity. COD samples were stored frozen until further processing. COD concentration was
measured by high temperature combustion using a Shimadzu TOC-5000 as described in Burdige and
Homstead (1994) [15].

2.3 Coagulation-flocculation experiments

All experiments were carried out on raw water collected from Medjerda dam (Tunisia) water quality
shown in Table 1. Standard jar tests were performed using a program-controlled jar test apparatus with six
flat impellers at a room temperature of 18.0–20.0 °C. The test water of 1.00 L was transferred into each 2.00
L plexiglass beakers. The initial pH was adjusted to the set value using 0.5 mol L-1 HCl and NaOH. The
solutions were stirred rapidly at 200 rpm for 1min during coagulant addition, followed by slow stirring at
40.0 rpm for 15.0 min and quiescent settling for 30.0 min. After deposition, water samples were collected
from 2.00 cm below the solution surface for subsequent measurements. The clarified water was extracted
from the beaker to measure chemical oxygen demand (COD) residual turbidity (Tu) and permanganate
oxygen demand ( Organic matter: OM).

2.4. Measurement of residual Al content

In general, residual Al ratio represents the residual Al induced by coagulant itself and it is an
important parameter of various coagulants in practical application [16-17].
The sample was acidified with nitric acid and was then analyzed. The Al concentration was analyzed using
atomic absorption spectrometry (AAS Nova 410, NF EN ISO 12020).

3. RESULTS AND DISCUSSION

3.1 General water characterization
The inorganic content of the Medjerda water is depicted in Table I. Water taken from
Medjerda dam has some high concentrations of chloride, sulfate and sodium ions. Nitrate, lead and
Manganese concentrations were very low. The pH is slightly basic.
 Mouna Jaouadi et al., J. Soc. Chim. Tunisie, 2013, 15, 175-181 177

Iron (Fe) and Silicium (Si) are the most prevalent inorganic constituent in Medjerda water. Calcium
(Ca) and sodium (Na) are the most abundant elements found in water.
Table I: Average characteristics of the raw water taken from Medjerda dam (Tunisia).
Table I: Average characteristics of the raw water taken from Medjerda dam (Tunisia).

Alkalinity KMnO4
Temperature Turbidity Conductivity
pH (TAC) oxidability
(°C) (NTU) (µS.cm-1)
(meq.L-1) (mgO2.L-1)
8.09 17. 5 4. 8 2560 128. 4 5.30
Chloride
Fe (mg.L-1) Mn (mg.L-1) Cu (mg.L-1) Al (mg.L-1) Zn (mg.L-1)
(mg.L-1)
116. 79 0. 32 0.001 0 0.017 0. 03
-1 -1 -1 -1 -1
Si (mg.L ) Hg (µg.L ) Na (mg.L ) K (mg.L ) Mg (mg.L ) Ca (mg.L-1)
2. 54 0.003 309.11 6. 05 61. 13 178.03
Nitrate Sulfate
Pb (µg.L-1) Ag (mg.L-1)
(mg.L-1) (mg.L-1)
3 204 0. 68 0.01

3.2 Residual concentration of aluminium

Figure 1: Variation of residual Al as a function of coagulant dose.

Residual aluminium concentration after sedimentation and filtration in the effluent must be
taken into consideration because of the health effects of aluminium. The studies on health effects of
aluminium in drinking water showed that high consumption of aluminium from drinking water may
be a risk factor for Alzheimer's disease and a relationship exists between the number of diagnosed
Alzheimer's disease cases and the level of aluminium present in the drinking water [18-19]. It has
been supposed that overdose of aluminium sulfate may result in high residual aluminium [20-21].
Residual Al decreased straightly within the dosage investigated for Al2(SO4)3. And there was very
little amount of residual Al existed in the purified water (Figure 1).
Residual aluminium concentration was affected by concentration of coagulant dosage. Kang et al.
[20] also noted that the highest soluble residual aluminium was found in the underdose aluminium-
based coagulant range in sample water containing low turbidity. Because the maximum
178 Mouna Jaouadi et al., J. Soc. Chim. Tunisie, 2013, 15, 175-181

contaminant level of aluminium in drinking water is 200 μg L−1, aluminium sulfate was found as an
effective and reliable coagulant due to required concentration and residual aluminium.

3.3 Optimization of coagulation-flocculation

3.3.1 Effects of water initial pH
In the coagulation-flocculation process it is very important to control pH. The aim of this
experiment was to investigate the influence of water initial pH on coagulation efficiency at the pre-
determined coagulant dosage and agitation speed of 40 mg.L -1 and 100 rpm respectively. Water
initial pH was adjusted in the range of 4 to 10 using 0.5 molL -1 HCl and NaOH solutions.
Figure 2 illustrates the effect of water initial pH on coagulation efficiency. The result shows that the
initial pH largely influenced both the organic matter (OM) and turbidity reduction.
The coagulant Al2(SO4)3 was a monomeric salt, it hydrolyzed as soon as added into water and it
would go through extremely rapid and uncontrolled hydrolysis when solution pH reached a certain
value [22-23].
With the increase of initial pH, the turbidity value decreased noticeably until the pH value of 6, but
increasing the pH above this point distinctly reduced the turbidity removal. Therefore, the optimum
pH for turbidity and OM reduction was 6.0.
Figure 3 shows that the optimum water initial pH for COD reduction was 5.0.Based on the above
consideration, the adopted optimum water initial pH was 6.0
200
16 16
190
14 OM 14
Tu 180
Turbidity (NTU)

12 12
OM( mgO2/L)

COD,ppm

170
10 10
160
8 8
150
6 6

4 4 140

2 2 130

2 3 4 5 6 7 8 2 3 4 5 6 7 8
pH pH

Figure 2: Variation of turbidity and OM as a function of pH. Figure 3: Variation of COD as a function of pH.

For aluminium hydroxide species at pH values between 4 and 6, the positively charged species are
pre-dominant, while the range 5.5–7.5 is the best for Al (OH) 3 precipitate [24-25].
A spectroscopic study [26] suggested that at low pH values (< 4.5), organic matter formed soluble
OM-Al complexes with A13+ , while at higher pH values (pH >5), adsorption of OM on the surface
of Al(OH)3, crystals becomes dominant, leading to a maximum removal of the OM. With
adsorption, the carboxylic group on a OM is first coordinated to the Al nuclei on the surface of Al
(OH)3, either by a ligand or by anion exchange; then further progressive adsorption of the rest of the
molecule on the surface of Al (OH) 3, is brought up by physical and chemical forces such as van der
Waals forces and H-bonding.
It can be seen that at pH 6 and above, the dominant mechanism was adsorption of organic species
on precipitated Al (OH) 3 at pH 5.0 and below, the mechanism involved precipitation of soluble
organic molecules by soluble cationic polymeric species; between pH 5 and 6, both mechanisms
may occur simultaneously.
 Mouna Jaouadi et al., J. Soc. Chim. Tunisie, 2013, 15, 175-181 179

Under different conditions, the different mechanisms or their combination may be dominant. As
indicated in the pH of raw, the effect of the initial pH on the particle and organic materials removal
was correlated with the coagulant hydrolyzates. For Al2(SO4)3, when pH was lower than 5.0, the
polymerization of Al3+ was inhibited at a certain degree and the primary Al speciation was the
positive monomer hydrolyzates, like Al(OH)2+, Al2(OH)24+, and Al3(OH)45+,etc. [27-28]. These
positive hydrolyzates were easy to neutralize the exterior negative charges of particle and organic
materials in Medjerda water, and further destabilize the colloids.
When pH was higher than 8.0, the suspension system was difficult to be destabilized because the
hydrolyzates were transformed to Al (OH)4 – [27-28].

3.3.2 Effects of coagulant dosage

In the coagulation flocculation process, coagulant dosage plays an important role in
determining the coagulation efficiency. The experiment was conducted at the pre-determined water
initial pH of 6 with an agitation speed of 100 rpm. The investigated coagulant dosages were in the
range of 5–30 mg.L-1.
Figure 4 shows the effects of coagulant dosage on OM (permanganat oxygen demand) and
turbidity. The results indicate that OM and turbidity value decreased substantially with the increase
of the coagulant dosage, .In the range of concentration used in this study (5-30 mg.L-1) the turbidity
and OM removal efficiency was achieved. The results also showed that at this range, the removal of
COD was maximum (Figure 5).
The higher removal of COD under relatively higher coagulant concentration (30 mg.L -1) could
result from the sweep flocculation process [29] due to the fact that the concentration was close to
the saturation level. Therefore, optimum coagulant dose was 30 mg.L -1.
Indeed, the effect of the colloidal particles concentration in water is important because they serve as
cores to the coagulation [30]. If the concentration of colloids is low, there are few particles to
ensure good flocculation, regardless their neutralization.
The coagulation of OM by hydrolyzing metal salts is generally described as co-precipitation, charge
neutralization and/or adsorption mechanisms, depending on the dose of coagulants and
concentration of dissolved organic carbon [31]. However, other phenomena may involve chemical
interactions between the metal ions and the functional groups of organic matter (MO) such as the
formation of complexes [32].

6.0 6.0 200

OM
5.5 Tu 5.5
180
5.0 5.0
Turbidity ( NTU)

OM( mgO2/L)

4.5 4.5 160

COD , ppm

4.0 4.0 140

3.5 3.5
120
3.0 3.0
2.5 2.5 100
2.0 2.0 80
1.5 1.5
5 10 15 20 25 30 0 5 10 15 20 25 30
Al2( SO4)3 dose,ppm Al2(SO4)3 dose,ppm

Figure 4: Effect of coagulant doses on permanganat Figure 5: Effect of coagulant doses

oxygen demand and turbidity removal. on COD removal.
180 Mouna Jaouadi et al., J. Soc. Chim. Tunisie, 2013, 15, 175-181

3.3.3 Effects of agitation speed

Agitation speed is one of the important factors in achieving higher flocculation efficiency
during the coagulation–flocculation process. In this experiment, the effects of agitation speed
between 20–100 rpm with rapid mixing time of 1 min on reduction of turbidity, OM and COD was
investigated. Figure 7 and 8 show the results from coagulation experiments carried out at the pre-
determined optimum water initial pH and coagulant dosage. Figure 6 shows that the turbidity and
OM value decreased with increasing agitation speed from 20 to 80 rpm. However, for a higher
agitation speed (80 rpm), the result indicated a substantial reduction in turbidity and OM removal
efficiency. Compared with the turbidity, there was a wider range of agitation speed that resulted in a
decrease of the OM removal efficiency. The figure 7 shows that the COD removal efficiency
decreased significantly between 40 and 80 rpm. Therefore, at a higher agitation speed, the
coagulation efficiency was not improved. However, at the lower agitation speed of 20 rpm (less
than 40 rpm), the flocs formed in coagulation-flocculation processes were easily broken and the
removal efficiency would decrease. Based on the actual measurement, the optimum agitation speed
was 80 rpm.

6.0 6.0 200

Tu
5.5 OM 5.5
180
5.0 5.0
Turbidity (NTU)
OM (mgO2/L)

160
COD,ppm

4.5 4.5

4.0 4.0
140

3.5 3.5 120

3.0 3.0
100
2.5 2.5
20 40 60 80 100
20 40 60 80 100
Agitation speed,rpm
Agitation Speed (rpm)

Figure 6: Effect of agitation speed Figure 7: Effect of agitation speed

on turbidity and OM removal. on COD removal.

CONCLUSION
Coagulation-flocculation using aluminium sulfate is a suitable process for the removal of
turbidity, organic matter and chemical oxygen demand when the operating parameters are
optimized.
The optimal operational conditions are as follows: initial pH value of 6, concentration of aluminium
sulfate of 30 mg.L-1 and agitation speed of 80 rpm. By applying these parameter values, maximal
turbidity, organic matter and chemical oxygen demand removal has been predicted and confirmed
experimentally. The obtained results are conform to legal Tunisian specification.

REFERENCES
[1] A. Bes-Pia, M.I. Iborra-Clar, A. Iborra-Clar, J.A. Mendoza-Roca, B. Cuartas-Uribe, M.I. Alcaina-Miranda,
Nanofiltration of textile industry wastewater using a physicochemical process as a pre-treatment, Desalination;
2005, 178 (1–3), 343.
[2] F. Kargi, M.Y. Pamukoglu, Adsorbent supplemented biological treatment of pretreated landfill leachate by fed-
batch operation, Bioresour. Technol; 2004, 94 (3), 285.
 Mouna Jaouadi et al., J. Soc. Chim. Tunisie, 2013, 15, 175-181 181

[3] J.E. Van Benschoten, J.K. Edzwald, Measuring aluminium during water treatment: methodology and application,
J. Am. Water Works Assoc; 1990, 71.
[4] P.T. Srinivasan, T. Viraraghavan, K.S. Surbramanian, Alumnium in drinking water: an overview, Water SA; 1999,
25 (1), 47.
[5] Directive 98/83/EC on the quality of water intended for human consumption, 1990.
[6] K. Derend, G. Van der Voet, W.H. Boer, Kidney Int; 2001, 59, 746.
[7] T.P. Flaten, Brain Res. Bull; 2001, 55, 187.
[8] X. Lu, Z. Chen, X. Yang, Water Research; 1999, 33, 3271.
[9] C. Gagnon, B.P.A. Grandjean, J. Thibault, Artif. Intell. Eng. 1997, 11, 401.
[10] S. Haydar, J.A. Aziz, Coagulation–flocculation studies of tannery wastewater using combination of alum with
cationic and anionic polymers, J. Hazard.Mater; 2009, 168 (2–3), 1035.
[11] A.W. Zularisam, A.F. Ismail, M.R. Salim, M. Sakinah, T. Matsuura, Application of coagulation-ultrafiltration
hybrid process for drinking water treatment: optimization of operating conditions using experimental design, Sep.
Purif. Technol; 2009, 65 (2), 193.
[12] M.I. Aguilar, J. Saez, M. Llorens, A. Soler, J.F. Ortuno, V. Meseguer, A. Fuentes, Improvement of coagulation–
flocculation process using anionic polyacrylamide as coagulant aid, Chemosphere; 2005, 58 (1), 47.
[13] F. AlMubaddal, K. AlRumaihi, A. Ajbar, Performance optimization of coagulation/ flocculation in the treatment of
wastewater from a polyvinyl chloride plant, J.Hazard. Mater; 2009, 161 (1), 431.
[14] V. Pallier, G. Feuillade-Cathalifaud, B. Serpaud, J.C. Bollinger, Effect of organic matter on arsenic removal during
coagulation/flocculation treatment, J. Colloid Interface Sci; 2009, 342 (1) 26.
[15] D.J. Burdige , J. Homstead, Fluxes of dissolved organic carbon from Chesapeake bay sediments, Geochimica
Cosmochimica acta; 1994, 58, 3407.
[16] GB 5476-85, Methods for pretreating ion exchange resins. National Standards of the People’s Republic of China.
[17] Y. Wu, W. Chen, Measurement of residual aluminium content in water treated by poly ferric-alumious-sulfate, J.
Guangxi Teachers Education University (Natural Science Edition) 2008, 25 (1) , 46–49 .
[18] D.R.C. McLachlan, C. Bergeron, J.E. Smith, D. Boomer, S.L. Rifat, Risk for neuropathologically confirmed
Alzheimer's disease and residual aluminium in municipal drinking water employing weighted residential histories,
Neurology; 1996, 46, 401.
[19] V. Rondeau, H. Jacqmin-Gadda, D. Commenges, C. Helmer, J.F. Dartigues, Aluminium and silica in drinking
water and the risk of Alzheimer's disease or cognitive decline: findings from 15-year follow-up of the PAQUID
cohort, Am. J. Epidemiol; 2009, 169, 489.
[20] M. Kang, H. Chen, Y. Sato, T. Kamei, Y. Magara, Rapid and economical indicator for evaluating arsenic removal
with minimum aluminium residual during coagulation process, Water Research; 2003, 37, 4599.
[21] J.E. Van Benschoten, J.K. Edzwald, Measuring aluminium during water treatment: methodology and application, J.
Am. Water Works Ass; 1990, 82, 71.
[22] W.P. Cheng, Comparison of hydrolysis/coagulation behavior of polymeric and monomeric iron coagulants in
humic acid solution, Chemosphere; 2002, 47, 963.
[23] M. Yan, D. Wang, J. Yu, J. Ni, M. Edwards, J. Qu, Enhanced coagulation with polyaluminium chlorides: role of
pH/Alkalinity and speciation, Chemosphere; 2008, 71, 1665.
[24] Parfitt, G.D, Principles of the Colloidal State,The Royal Institute of Chemistry, England. 1967.
[25] M.I. Aguilar, M. Llor!ens, J. Sáez, L.M Leal, J.F. Ortuno, J.J Torres, Treatment of slaughterhouse wastewater by
coagulation–flocculation. Influence of pH and coagulant aids. Anales de Quımica Int. Ed., 1998, 94, 231.
[26] X.Q. Lu, Z.L. Chen and X.H. Yang, Spectroscopic study of aluminium speciation in removing humic substances
by Al coagulation, Water Research; 1999, 33(15), 3271.
[27] P. Zhang, Z. Wu, G. Zhang, G. Zeng, H. Zhang, J. Li, X. Song, J. Dong, Coagulation and characteristics of
polyaluminium chlorides PAC–Al30 on humic acid removal from water, Sep. Purif. Technol; 2008, 63, 642–647.
[28] B. Corain, G.G. Bombi, A. Tapparo, M. Perazzolo, P. Zatta, Aluminium toxicity and metal speciation: established
data and open questions, Coordin. Chem. Rev; 1996, 149, 11–22
[29] J.L. Lin, C. Huang, C.J.M. Chin, J.R. Pan, Coagulation dynamics of fractal flocs induced by enmeshment and
electrostatic patch mechanisms, Water Research; 2008, 42, 4457.
[30] J. Roussy, M. Van Vouren, E. Guibal, Influence of chitosan characteristics on coagulation and flocculation of
bentonite suspensions. Water Research; 2005, 39, 3247.
[31] G.A. Edwards, A. Amirtharajah, Removing color caused by humic acids, J. AWWA; 1985, 77(3), 50.
[32] V.L. Snoeyink , D. Jenkins, Water Chemistry, Wiley, New York, 1980.