The Islamic University of Gaza- Environmental Engineering Department

Water Treatment (EENV- 4331)

Lecture 5: Coagulation and Flocculation

Prepared by
Husam Al-Najar

Colour in water

Definitions
Coagulation is a water treatment process that causes very small
suspended particles to attract to one another and form larger
particles.
consist of adding a floc-forming chemical reagent to a water to
enmesh (catch) or combine with non-settleable colloidal solids
and slow-settling suspended solids to produce a rapid-settling
floc.
Flocculation is a water treatment process following coagulation,
which uses gentle stirring to bring the suspended particles
together so they will form larger more settleable clumps called
floc.

Many of the contaminants in water contain matter in the

colloidal form.
These colloids result in a stable suspension.
In general the suspension is stable enough so that gravity
forces will not cause precipitation of these colloidal particles.
So they need special treatment to remove them from the
aqueous phase. This destabilization of colloids is called
coagulation.

Settling time of particles with a density of 2,650 kg/m3

Diameter (m)

Type of particles

Settling times over

30 cm

1.10-2

gravel

1.10-3

Coarse sand

3 sec

1.10-4

Fine sand

38 sec

1.10-5

silt

33 min

1.10-6

bacteria

35 hours

1.10-7

clay

230 days

1.10-8

colloids

63 years

0.3 sec

Principles
In surface water particles difference in sizes are present.
A common classification is:
Molecules sizes smaller than 1 nm
Colloids generally with dimensions between 1 nm and 1 m
Suspended matter having sizes larger than 1 m.

colloids: commonly present in surface water are humic acids, proteins,

colloidal clay, silica and viruses.
suspended matter: Bacteria, algae, silt, sand and organic debris.
Suspended matter - when it is larger than 5-10 m can be removed quite
easily by filtration or sedimentation and filtration.
The removal of colloids is possible by slow filtration in cases the water is
not strongly polluted.

There are several possible origins of primary surface charge:

1. Adsorption of potential determining ions.

This involves preferential adsorption of a specific type of ion on

the surface of the colloid.

Adsorption is usually van derWaals or hydrogen bonding.

For example, a surfactant on clay surface, humic

acid on silica, OH- on many minerals.

The charge that results is a function of concentration and type of

ion in solution, pH, etc.)

2. Lattice imperfections or isomorphic replacement. This is very

common in clay minerals. For example, the isomorphic replacement
of Al3+ for Si4+ as shown below.

3. Ionogenic groups at surface. (ionizable surface groups such as

carboxyl, amino, hydroxyl, sulfonic, etc.) The charge here is usually
dependent on pH.

Electrical Double Layer (EDL):

If we put a charged particle in a suspension with ions, then the
primary charge will attract counter ions (opposite charged ions) by
electrostatic attraction.
The primary charge cannot attract an equal amount of counter
charge because a gradient of counterions is established that drives
the counterions away from the surface.
The formation of the electrical double layer (EDL) then occurs via
attraction of oppositely charged counterions by the primary
surface charge and then a diffusion of the counterions away from
the surface.

The counterions are mobile, the primary charge is not.

The EDL development is schematically

shown here:

When colloids are subjected to an electrical field they will migrate

generally toward the positive electrode of the field .
They move because the inner part of the colloid (with higher charge
density than the overall colloid) will respond to the field and leave the
outer diffuse layer behind.
The EDL actually shears at a plane and the potential (voltage) of
the EDL at this shear plane is called the Zeta Potential, z.

The zeta potential represents the net charge between the primary
charge and the counter charge in the EDL located between the
surface and the shear plane.
Its with this charge that the colloid interacts with other colloids.

As a result of this EDL there is

a net electrostatic
repulsion/attraction developed
between colloids.

The net resultant force is a result of:

1. attractive potential energy (mostly van der Waals forces), Va.
These forces are very strong at short separation distances

Va 1
r6
2. repulsion potential energy (electrostatic forces), VR. (by
Coulombs law).

VR 1
r2

Zeta Potential Model

The magnitude of these forces is measured by the zeta potential,
which is:

where:
Z is the zeta potential,
q is the charge per unit area,
d is the thickness of the effective charge layer, and
D is the dielectric constant of the liquid.
The greater the zeta potential, the greater are the repulsion forces
between the colloids and, therefore, the more stable is the colloidal
suspension.

The interactions are:

The reduction of the zeta potential to a degree where the attractive van der
Waals forces and the agitation provided cause the particles to combine;
The aggregation of particles by interparticulate bridging between reactive
groups on the colloids;
The enmeshment of particles in the precipitate floc that is formed.

Chemicals reduce colloidal surface charge and form

precipitates that enhance the clustering process and
sedimentation.
Colloids will not settle or filter until they agglomerate to
a larger size.

Briefly: There are two major forces acting on colloids:

1) electrostatic repulsion (simply, negative colloids repel other
negatively charged colloids)

2) intermolecular, or van der Waals, attraction.

Coagulants can be used to reduce the electrostatic repulsive

forces

The electrostatic repulsion reduced by the addition of

countercharged ions [Al3+]

Chemical coagulation:
The process in which chemicals like Al or Fe salts added to the
water to transform the impurities after reaction with the hydrolyzing
salts into large flocs, which can be removed easily by
sedimentation and filtration.

Lowering of the charge responsible - if high enough - for the net

repulsive force will lead to a situation where collisions can occur
resulting in agglomeration.

Coagulation is defined - in a stricter sense - as the

destabilization of colloids.

In waters with higher ionic strengths the colloids are already

present in a destabilized form.
Destabilization here has been caused by the mono- and divalent
ions present in the water (electrostatic coagulation).

Coagulants are added here to improve the flocculation kinetics.

After coagulation the destabilized particles can collide, aggregate

so flocs can be formed. This step is called flocculation.

Flocculation: The process of agglomeration of the destabilized

particles to such a size that separation by sedimentation and
filtration is possible.
In flocculation one can make a distinction between perikinetic and
orthokinetic flocculation. Brownian motion is the driving force in the
agglomeration of destabilized particles up to m-level perikinetic
flocculation).
Above ~ 1 m the influence of Brownian motion on the collision rate
of the particles can be neglected, then artificial mixing is necessary
to get an efficient flocculation. That part of the flocculation process
is called orthokinetic flocculation.

In water treatment two main types of coagulants are used i.e.

aluminum and iron salts.
Type of coagulant

formula

most common
form

reaction
with water

aluminum sulfate

Al2 (SO4)3. 14-18 H2O

lumps or powder

acidic

Sodium aluminate

NaAlO2 or Na2Al2O4

Powder

alkaline

Poly-aluminiumchloride Aln(OH)mCl3n-m

Solution or powder

acidic

Ferric sulfate

Fe2(SO4)3.9H2O

Small crystals

acidic

Ferris chloride

FeCl3. 6H2O

Lumps or solution

acidic

Ferrous sulfate

FeSO4. 7H2O

Small crystals

acidic

Before application one must always prepare a solution of the coagulants.

In that solution the coagulant is most effective.

if it is present in the trivalent state i.e. Fe3+ or At 3+ resulting in a pH of <

1.5 of the solution.
When a solution of alum is added to the water to be treated the following
reactions take place.
Hydrolysis reaction: Al 3+ + 3H2O

Al(OH)3 + 3H+

If enough alkalinity is present the following reactions can occur:

If CO32- is present: CO32- + H+
or with HCO3- : HCO3- + H+

HCO3- + H2O
CO2 + H2O resulting in a lowering of pH

The compounds formed are positively charged and can

interact with the impurities like colloids.
The solubility of Al(OH)3, is very low. So precipitation occur in
the form of flocs.

As H+ - and OH- -ions are involved the pH of the water during

coagulation is important for the charge of the hydrolysis
products.
The chemical composition of the water is important too
because divalent ions like SO42- and HPO42- can replace the
OH- -ions in the complex and hence can influence the
properties of the precipitate

[Al (H2O)6)3+

[Al (H2O)5OH]2+ + H

[Al (H2O)5OH]2+

[Al(H2O)4 (OH)2]+ + H+

[Al(H2O)4 (OH)2]+

[Al(H2O)3(OH)3] + H+
precipitate

[Al(H2O)3(OH)3]

[Al(H2O)2(OH)4]- + H+
soluble

If enough alkalinity is present the H+-ions formed will be taken away

and the precipitate [Al(H2O)3(OH)3] or just Al(OH)3 will be formed.
At pH-values higher than 7.8 aluminates-ions [Al(H2O)2(OH)4]- or just
Al(OH)4- are formed which are negatively charged and soluble in water.
In coagulation with an aluminum compound pH-values higher than
7.8 must be avoided for that reason.

For iron compounds the same type of hydrolysis reactions take place:
Fe3+ + 3H2O

Fe(OH)3 + 3H+

followed by the reaction of the H+ -ions with the alkalinity.

Here there exist the hydrated ferric-ion [Fe(H2O)6]3+ with the same
hydrolysis equations as for [Al(H2O)6] 3+.

The formation of [Fe(H2O)2(OH)4]- or Fe(OH)4occurs only at high pHvalues not common in conventional treatment, so the upper pH-limit for
coagulation with Fe3+ is much higher than for Al3+ i.e. 10 instead of 7.8
(Al3+).
Iron compounds have less tendency to form polynuclear species
compared with aluminium.

The efficiency of the coagulation process is dependent upon a number of

parameters like:

The water quality

The type and concentration of colloids (part of water quality)
The type and dose of coagulant
The value of pH during coagulation

optimal pH-range for Al: 6 - 7.8

Fe3+: 4 - 10
Fe2+: > 8.5
Temperature
Mixing conditions
In practice the coagulation process involves a rapid mixing, flocculation,
sedimentation and filtration step. Some of these steps are sometimes
combined in water treatment practice.

JAR TESTS

The selection of a coagulant requires the use of laboratory or pilot plant

coagulation studies
Usually laboratory studies using the jar test are adequate for selecting a
coagulant for a water treatment plant
A jar test is usually used to determine the proper coagulant and coagulant
aid, if needed, and the chemical dosages required for the coagulation of a
particular water.
Samples of the water are poured into a series of containers
Various dosages of the coagulant and coagulant aid are added
The contents are rapidly stirred to simulate rapid mixing
Then the contents are gently stirred to simulate flocculation
After a given time, the stirring is ceased and the floc formed is
allowed to settle.

The most important aspects to note are:

The time for floc formation,
The floc size,
Its settling characteristics,
The percent turbidity and color removed, and
The final pH of the coagulated and settled water.
The chemical dosage determined from the procedure gives an
estimate of the dosage required for the treatment plant.

The jar test simulate the coagulation/flocculation process in a batch

mode. A series of batch tests are run in which pH, coagulant type
and dosage and coagulant aid are varied to get the optimal dosage
(lowest residual turbidity).
An economic analysis is performed to select these parameters.
Jar tests generally are performed using 6 one-liter samples of the
water to be treated.
To these samples a range of coagulant (and possibly coagulant aid)
dose is added (one sample is usually a blank).

Immediately after the coagulant is added the samples are "flash

mixed" for approximately one minute.
The stirrer speed is then reduced to simulate a flocculation basin.
Flocculation mode is generally maintained for about 20 minutes.
At the end of the flocculation period the stirrers are turned off and
the floc is allowed to settle for one-half hour.

After this settling period supernatant samples are drawn off from
each sample and analyzed for turbidity and sometimes alkalinity and
pH.

Jar Test

Typical results from a jar test series might look like:

Dose (mg/L)

Results of jar-test experiment with

varying coagulant dosing

Results of jar-test experiment of

with varying pH

Coagulation and Flocculation Process Design

P
G
V
Where,
G = velocity gradient, s-1
P = power input, W
V = volume of water in mixing tank, m3
= dynamic viscosity, Pa.s
Viscosity is a measure of the resistance of a fluid which is being deformed by either shear stress or
tensile stress. The SI physical unit of dynamic viscosity is the pascal-second (Pas), which is identical to
Nm2s

velocity gradient With respect to material being sheared, velocity gradient is the change
dv in relative velocity v between parallel planes with respect to the change dr in
perpendicular distance r throughout the depth of the material. Velocity gradient has the
same dimensions as rate of shear, which is reciprocal seconds.

Rapid Mix
Types of Rapid Mixing Systems:
Mechanical devices in a dedicated basin (Most common
application)
In-Line blenders (Efficient, Effective at all flows)
Hydraulic methods (Simple, effective, non-mechanical.
Energy may vary with flow)

Air Mixing (Simple, advantages if aeration is required. Not

common. May cause scum and floatable debris)

Mechanical Flash Mixers: Propeller Type

Mechanical Flash Mixers: Turbine Type

In-Line Blenders: Mechanical Type

In-Line Blenders: Static Type

Rapid Mix Design Considerations

Most chemical reactions in coagulation occur in less than 0.1 seconds.
It is therefore important that mixing be as instantaneous and complete as
possible.
Since G is a measure of the mixing intensity, selection of G and Gt where t
is the detention time of the fluid in the rapid mixer, are important parameters
in determining the proper mixing for both rapid mix and flocculation reactor
design.
Coagulation occurs in two ways:
1)

2)

By adsorption of soluble hydrolysis coagulant species on the colloid

particles and destabilization by charge neutralization. These reactions
occur in about 1 second.
Sweep floc coagulation where the coagulant exceeds its solubility limit
and precipitates and traps the colloid particles. Sweep floc coagulation
occurs in the range of 1 to 7 seconds.

Jar tests can be used to determine which coagulation mechanism is

predominant:

Charge reversal is predominant if dose-turbidity curves are parabolic in

nature and have a minimum point. For this case G values in the range
of 3,000 to 5,000 s-1 and detention times of about 0.5 s are
recommended (1500 < Gt < 2,500). In-line blenders can be used.
If the dose response curve does not show charge reversal (is relatively flat
at higher doses), then the predominant mechanism is sweep floc
coagulation. For this case G values in the range of 600 to 1,000 s-1 and
detention times of about 1 to 10 s are recommended (600 < Gt <
10,000).
In the softening process, for dissolution of CaO to Ca(OH)2, mixtures,
detention times of 5 to 10 minutes are usually required. G values are
around 700 s-1 and in-line blenders are typically not used.

Rapid Mix Tank Design

Maximum Tank Volume = 8 m3 (due to mixing equipment and geometry
constraints)
Mixing Equipment: Electric Motor, gear-type speed reducer, turbine of
axial shaft impeller.

Usually the turbine impeller provides more turbulence and is preferred in

rapid mix tanks.
The tanks are usually, baffled horizontally into two or three compartments
in-order to provide sufficient residence time.
Tanks should also be vertically baffled to minimize vortexing.
Chemicals should be added below the impeller, point of most mixing.

Mixing impeller types used in water treatment

Rapid Mix Tank

Rapid Mix Tank Design Power Requirements

For the design of Rapid mix and flocculation tanks, the power imparted to
the liquid in a baffled tank by an impeller may be described by the
following equation for fully turbulent flow:
3
5

P K T n Di

P = power, W
KT = impeller constant
n = rotational speed, revolutions/s
Di = impeller diameter, m
= density of liquid, kg/m 3
For un-baffled tanks, the power imparted may be as low as one-sixth of
that predicted by this equation.

The impeller constant, KT, can be obtained from the following table:
Values of the impeller constant, KT
Type of impeller

KT

Propeller, pitch of 1, 3 blades

0.32

Propeller, pitch of 2, 3 blades

1.00

Turbine, 6 flat blades, vaned disc

6.30

Turbine, 6 curved blades

4.80

Fan turbine, 6 curved blades at 45o

1.65

Shrouded turbine, 6 curved blades

1.08

Shrouded turbine, with stator, no baffles

1.12

Common Rules of Thumb:

Design Liquid depth = 0.5 to 1.1 times the basin diameter or width.
Impeller diameter is between 0.3 and 0.50 times the tank diameter or
width.
Vertical baffles extend into the tank about 10 % of the tank diameter or
width.

Impellers typically do not exceed 1.0 meter in diameter.

Liquid depth may be increased to between 1.1 and 1.6 times the tank
diameter if dual impellers on the shaft are employed. When dual impellers
are employed, they are spaced about two impeller diameters apart.
Transfer efficiency of motor power to water power is about 0.8 for a single
impeller.

EXAMPLE OF RAPID MIX TANK DESIGN

A city is planning for the installation of a water treatment plant to remove
iron. A low-turbidity iron coagulation plant has been proposed with the
following design parameters:
Q = 2 m3/s
Rapid mix detention time, t = 10 s
Rapid mix G = 1,000 s-1
Design a rapid-mix basin and size the mixing equipment.

Solution:
The volume of the rapid-mix tank by is:

V Q t 2 m3 / s 10 s 20m3
Since the minimum tank volume is 8 m3 is a guideline, tanks in parallel will
have to be provided.
The design is also constrained by the availability of mixers and those
limitations need to be evaluated.

Assume the following mixers are available:

Model

Rotational
speeds, rpm

Power,
kW

Model

Rotational
speeds, rpm

Power,
kW

JTQ25

30,45

0.18

JTQ300

110,175

2.24

JTQ50

30,45

0.37

JTQ500

110,175

3.74

JTQ75

45,70

0.56

JTQ750

110,175

5.59

JTQ100

45,110

0.75

JTQ1000

110,175

7.46

JTQ150

45,110

1.12

JTQ1500

110,175

11.19

JTQ200

70,110

1.5

rpm = revolutions per minute

JTQ models have variable speeds from 1-45 rpm

The largest available mixer can achieve a water power of:

11.19kW 0.8

8.95 kW

The 0.8 is the assumed efficiency of transfer of motor power to water

power. Given a G of 1,000 s-1, and a viscosity at 18oC as 1.053x10-3 Pas,
the required mixing can be calculated as:

P
G
V
P
8.95 103 W
3
V 2

8.50
m
G 1,000 s1 2 1.053 103 Pa s

This means that using the largest available mixer from the manufacturer, to
supply 20 m3 volume requirement the number of tanks needed are:

Number
total tank volume required
of

Tanks volume requirement per tank based on mixing

20m3

2.35 or 3 rapid mix tanks

3
8.50m / tank
The volume for each tank is 6.67 m3. For the JTQ1500 the rotational speed
is 110 rpm (1.83 rps) and a turbine with six flat blades (vaned disc), the
impeller diameter can be estimated as:
3
5

P K T n Di

Di
K n 3
T

15

0.23

15

8.95 103 W

6.30 1.83rps 1.053 10 3 Pa s

0.75 m

Using a ratio of impeller diameter to tank diameter of 0.33, the tank diameter
would be equal to:

impeller diameter
Tank

diameter ratio of impeller diameter to tank diameter

0.75m
2.27 m
0.33

The surface area of the tank would be equal to:

2
Tank

2
surface 2.27 m 4.05 m
area
4

With three 6.67 m3 tanks, the depth of each tank would be:

3
tank
volume
6.67
m
Tank

1.65 m
2
depth
tank area
4.05 m

Need to check the liquid-depth to tank ratio:

liquid depth 1.65 m

0.73

tank diameter 2.27 m

This is within the guideline of 0.5 to 1.1.

FLOCCULATION
Flocculation is the most important parameter impacting particle removal
efficiency.
The primary objective of flocculation is to bring particles in contact with one
another so they will collide,stick together, and grow to a size that will easily
settle.
The right amount of mixing will enable flocculation to occur without settling
the particles out in the floc basin or shearing the particles due to over
mixing.
The mixing for flocculation must be controlled in a narrow range and must
be able to vary G by a factor of two or three.

The following table illustrates the required mixing for different types of floc
particles encountered in water treatment

Table 1. Gt values for flocculation

Type

G (s-1)

Gt (unitless)

Low-turbidity, color
removal coagulation

20 70

60,000 to 200,000

High-turbidity, solids
removal coagulation

30 - 80

36,000 to 96,000

Softening, 10% solids

130 - 200

200,000 to 250,000

Softening, 39% solids

150 - 300

390,000 to 400,000

Types of Flocculation Tanks

Mechanical Flocculators

Flocculation Practice in Water Treatment

Hydraulic Flocculators

The axial flow flocculators are typically used because they impart a nearly
constant G in each compartment.
Flocculators are designed to have a minimum of three compartments to
provide for tapered mixing.
The velocity gradient, G is tapered so that it is larger in the first
compartment and less is the other compartments as the floc grows.

Table (1) provides average values of G for the flocculation basins.

Some common rules of thumb for axial flow impellers are that the diameter
of the impeller is between 0.2 and 0.5 times the width of the chamber and
that the maximum impeller diameter is about 3 m.

FLOCCULATION POWER REQUIREMENTS

PADDLE FLOCCULATORS:
The following equation can be used to calculate the power requirements for
a paddle mixer:

CD A v p

P = power imparted, kW
CD = paddles drag coefficient
= density of fluid, kg/m 3
A = cross sectional area of paddles, m 2
vp = relative velocity of paddles with respect to fluid, m/s

Rules of thumb:
Paddle tip speed = 0.1 to 1.0 m/s
Relative velocity of the paddles to the fluid = = 0.6 to 0.75 the paddles tip
speed.
Drag coefficient vary with length to width ratio.

Total paddle-blade area on a horizontal shaft not to exceed 15 to 20

percent of the total basin cross sectional area to avoid excessive rotational
flow.

PNEUMATIC MIXER:

h 10.33
P KQa ln

10.33
P = power imparted, kW
Qa = air flow rate at atmospheric pressure, m 3/min
K = constant = 1.689
h = air pressure at the point of discharge, m

HYDRAULIC MIXER:
The following expression can be used for a hydraulic mixer:

P Qh
P = power imparted, kW
Q = fluid flow rate, m3/s
= specific weight of fluid, kN/m 3
h = head loss through the mixer, m
The specific weight of water is equal to the product of the density and the
acceleration due to gravity. Typically at normal temperatures the specific
weight of water is taken to be 9.81 kN/m3

Upflow Solids-Contact Basin

An upflow solids contact basin can be used to provide, mixing, flocculation
and sedimentation/clarification all in one tank as shown below.

EXAMPLE OF FLOCCULATION TANK DESIGN

A city is planning for the installation of a water treatment plant to remove
iron. A low-turbidity iron coagulation plant has been proposed with the
following design parameters:
Q = 2 m3/s
Flocculation t = 20 minutes
Flocculation G = 30 s-1
T = 18oC
Design a flocculation system and size the mixing equipment.

The volume of the flocculation basin is:

V Q t 2m3 / s 20min 60s / min 2,400m3

Need to divide the basin into three compartments. Each compartment will
have the following volume:

3
total
volume
2,400
Compartment

800m3
volume
3 compartments
3

Since the average G is 30 s-1, we will want to taper the mixing intensity such
that it is reduced by 50% from the first compartment to the last compartment.
The design G for each compartment will be 40 s-1, 30 s-1 and 20 s-1 to yield
an average of 30 s-1.

Following the same approach for the rapid mix system, we calculate the
average maximum volume that can be mixed.

P
8.95 103 W
3
V 2

9,444
m
G 30 s1 2 1.053 103 Pa s
For this G value the mixer power will not be limiting. Using 800 m3 tank and
dividing it into three 267 m3 compartments, the power required for the first
compartment would be:

P1 G V 40 s 1.053 103 Pa s 267m3

450 W 0.450kW
2
1

1 2

With 80 percent efficiency, the motor power should be:

P1 motor 0.450kW / 0.8 0.5625 kW

The variable speed motors of JTQ75 should be evaluated.

Assuming a typical water depth of a flocculator of 4 m, and a square tank

arrangement, the dimensions of the tank compartment would be:

3
Top volume 267 m 66.75m2
area
depth
4m

and

Sides 66.75m2

12

8.17m

Using a diameter-to-width ratio of 0.3, the impeller diameter would be:

Impeller 8.17 m 0.3 2.45m

diameter

Using a three-blade propeller with a pitch of 2, the required rotational speed

would be:
3

570 W 0.8

5
5

K T Di 1.00 2.45m 1,000kg / m3

P

13

0.005166 0.173 rps 10.4 rpm

This is within the variable speed range of 1 to 45 rpm for the JTQ-F motor.