5
Dissolved Air Flotation (DAF)
for Wastewater Treatment
Puganeshwary Palaniandy, Hj. Mohd Nordin Adlan,
Hamidi Abdul Aziz, and Mohamad Fared Murshed
Citation Information
Waste Treatment in the Service and Utility Industries
Edited by Yung - Tse Hung, Lawrence K . Wang, Mu - Hao
Sung Wang, Nazih K . Shammas and Jiaping . Paul Chen
University 6000 Broken Sound Parkway NW, Suite 300, Boca Raton, FL
33487–2742 CRC Press 2017
Print ISBN: 978-1-4200-7237-2
eBook ISBN: 978-1-351-67341-9
https://doi.org/10.1201/9781315164199
Universiti Sains Malaysia
Yung-Tse Hung
Cleveland State
CONTENTS
5.1
5.2
I ntroduction........................................................................................................................... 146
Types of Flotation.................................................................................................................. 146
5.2.1 Electroflotation.......................................................................................................... 146
5.2.2 Dispersed Air Flotation............................................................................................. 147
5.2.3 Dissolved Air Flotation............................................................................................. 147
5.3 Process Description of DAF.................................................................................................. 148
5.4 Theory of DAF...................................................................................................................... 150
5.4.1 Bubble Formation...................................................................................................... 151
5.4.2 Bubble–Particle Attachment...................................................................................... 151
5.4.3 Flotation of Bubble–Particle Agglomerate................................................................ 151
5.4.4 K inetics of Flotation.................................................................................................. 151
5.4.5 Solubility of Air......................................................................................................... 157
5.4.6 Bubble Generation..................................................................................................... 159
5.4.7 Collision..................................................................................................................... 160
5.4.8 Interception and Diffusion......................................................................................... 163
5.4.9 Tank Design............................................................................................................... 166
5.5 Advantages of DAF Application in Wastewater Treatment................................................... 170
5.6 Application of DAF Process in Wastewater Treatment......................................................... 170
5.7 Application of DAF Process in Landfill Leachate Treatment............................................... 173
List of Nomenclature...................................................................................................................... 176
References....................................................................................................................................... 178
Abstract
Flotation system consists of four major components—air supply, pressurizing pump, retention tank, and flotation chamber. The theory of dissolved air flotation (DAF) process is to
separate suspended particles from liquids by bringing the particles to the surface of the liquid.
In most cases, DAF is an alternative process to sedimentation and offers several advantages,
including better final water quality, rapid startup, higher rates of operation, and thicker sludge.
Additionally, DAF systems need less space compared with normal clarifiers, and due to their
modular components, they allow easy installation and setup. This chapter covers types of
flotation, process description of DAF, theory of DAF, advantages of DAF application in
wastewater treatment, application of DAF process in wastewater treatment, and application of
DAF in landfill leachate treatment.
145
146
Waste Treatment in the Service and Utility Industries
Downloaded by [Hamidi Aziz] at 16:04 26 September 2017
5.1 INTRODUCTION
There are various types of flotation process available for different applications. The technology has
been applied in many industries such as in mineral processing [1,2], wastewater clarification [3–6],
artificial recharge [7], and potable water treatment [8,9]. Basically, flotation is a process of separating solids from a body of liquid by using air bubbles.
Flotation has been used in the mining and chemical processing industries for over 100 years [10].
However, the history of flotation goes back even earlier. The ancient Greeks used this process over
2000 years ago to separate minerals from gangue [11]. The development of the process to its current
modern practices took several years. According to Kitchener, Haynes was able to separate minerals
using oil in 1860 [12]. His method was patented. In 1905, Salman, Picard, and Ballot developed a
process to separate sulfate grains from water by adding air bubbles and a small amount of oil to
enhance the process. This was called “froth flotation.” In 1910, T. Hoover developed the first flotation machine, which was not much different from the current equipment. A few years later, in 1914,
Callow introduced a new process called “foam flotation” [12]. This process involved the introduction of air bubbles through submerged porous media. In fact, froth and foam flotation processes are
generally known as dispersed air flotation processes and are used widely in the mineral processing
industry. The development of the electrolytic flotation process can be traced back to 1904. The
process was suggested by Elmore brothers who showed that electrolysis could produce bubbles for
flotation. It was not used commercially at that time.
Dissolved air flotation was patented in 1924 by Niels Peterson and Carl Sveen in Scandinavia
[13]. It was initially used to recover fibers and white water in the paper industry. The use of dissolved air flotation (DAF) in the treatment of wastewater and potable water began in the late 1960s.
Edzwald and Walsh reported that DAF has been used for water clarification in Europe especially
in the Scandinavian countries for more than 20 years, [10]. Heinanen, in his survey on the use of
flotation in Finland, indicated that the first DAF plant for potable water clarification was constructed
in 1965, and that by 1988, there were 34 plants in operation [14]. However the first application of
flotation for a water reclamation plant was introduced in the early 1960s in South Africa [15].
In the United Kingdom, the first full-scale water treatment plant using this process was commissioned in 1976 at the Glendye Treatment Works of the Grampian Regional Council, Scotland [16].
Experiments carried out by researchers at the Water Research Centre showed that flotation is a more
rapid method of solid–liquid separation than sedimentation [17].
5.2 TYPES OF FLOTATION
The basic idea in the flotation process is the solid–liquid separation process using bubbles. The
bubbles are produced using any gas that is not highly soluble in liquid. However, in actual practice,
air is the gas most commonly used because it is easily accessible, safe to use, and less expensive.
The method of producing bubbles gives rise to different types of flotation processes, namely, electrolytic flotation, dispersed air flotation, and dissolved air flotation [18,19].
5.2.1 Electroflotation
Electrolytic flotation is also known as electroflotation. The bubbles are produced by passing a
direct current between two electrodes and by generating oxygen and hydrogen in a diluted aqueous solution. Bubbles produced from electrolytic flotation are smaller compared with those produced from dispersed air flotation and DAF. Thus, this process is favorable for the removal of
low-density fragile flocs. This process is suitable for effluent treatment [20], sludge thickening,
and water treatment installations of 10–20 m3/h. Figure 5.1 shows a typical arrangement of an
electroflotation tank.
147
Dissolved Air Flotation (DAF) for Wastewater Treatment
Sludge removal belt
Water level
Sludge trough
Raw water
Downloaded by [Hamidi Aziz] at 16:04 26 September 2017
Clarified
water
effluent
Electrodes
FIGURE 5.1 Electroflotation tank. (From Adlan, M.N., A study of dissolved air flotation tank design variables and separation zone performance. PhD thesis, University of Newcastle Upon Tyne, 1998.)
5.2.2 Dispersed Air Flotation
Dispersed air flotation has two different systems to generate bubbles, namely, foam flotation and
froth flotation. In the foam flotation system, bubbles are generated by forcing air through a porous
media made of ceramic, plastic, or sintered metal [21]. Figure 5.2 shows a typical arrangement for
bubble generation through a medium or diffuser.
In the froth flotation system (shown in Figure 5.3), a high-speed impeller or turbine blade rotating
in the solution is used to produce air bubbles.
Dispersed air flotation normally produces large air bubbles measuring >1 mm in diameter [22].
It is used mainly for the separation of minerals and removal of hydrophobic materials such as fat
emulsions in selected wastewater treatment. This process was assessed for potable water treatment
but was found unsuitable [23].
5.2.3 Dissolved Air Flotation
The bubbles in DAF are produced by the reduction in pressure of a water stream saturated with
air. The three types of DAF are vacuum flotation, microflotation, and pressure flotation, of which
Influent
Foam
concentrate
Air
Effluent
Diffuser
FIGURE 5.2 Foam flotation. (From Adlan, M.N., A study of dissolved air flotation tank design variables and
separation zone performance. PhD thesis, University of Newcastle Upon Tyne, 1998.)
148
Waste Treatment in the Service and Utility Industries
Air
Mineral/water
slurry
Foam
trough
High speed impeller
FIGURE 5.3 Froth flotation. (From Adlan, M.N., A study of dissolved air flotation tank design variables and
separation zone performance. PhD thesis, University of Newcastle Upon Tyne, 1998.)
Downloaded by [Hamidi Aziz] at 16:04 26 September 2017
Sludge
Air
Influent
Flotation
chamber
Clarified
effluent
Flocculating agent
(if required)
FIGURE 5.4 Full flow pressure operational modes. (From Adlan, M.N., A study of dissolved air flotation
tank design variables and separation zone performance. PhD thesis, University of Newcastle Upon Tyne, 1998.)
the pressure flotation is the most important and widely used in water and wastewater treatments. In
the pressure flotation, the air is dissolved in water under high pressure and released at atmospheric
pressure through a needle valve or nozzle, which produces small air bubbles.
There are three kinds of pressures dissolved in air flotation that can be used: full-flow, spilt-flow,
and recycle flow pressure flotation [24,25].
• Full flow: The entire influent is pressurized and then released in the flotation tank, where
the bubbles are formed. This type of flotation is used for influents that do not need flocculation but require large volumes of air bubbles (Figure 5.4).
• Spilt flow: A part of the influent is pressurized, and the remaining flows directly into the
flocculation or flotation tank. This type of flow is cost-effective compared with full-flow
pressure flotation, because the saturator and the feed pump handle only a portion of the
total flow, thus requiring a smaller saturator and feed pump. However, split flow provides
less air in the system. As a result, it has to be operated at high pressure to provide the
same amount of air. This type of flow is used for influents containing suspended particles
susceptible to the shearing effects of a pressure pump. It is also suitable for influents containing suspended particles at low concentration, due to low air requirement (Figure 5.5).
• Recycle flow: The influent flows into the flocculation or flotation tank if flocculation process is
not required. A portion of the treated influent is recycled, pressurized, saturated with air, and
released to the flotation tank. This type of flow is applied to influents that need coagulation and
flocculation. It is a common type of flow, and it is used more often than other types (Figure 5.6).
5.3 PROCESS DESCRIPTION OF DAF
The DAF system consists of air and water supply, saturator, and flotation chamber or tank. There are
two types of flotation tanks: circular tank and rectangular tank. Wastewater treatment and sludge
149
Dissolved Air Flotation (DAF) for Wastewater Treatment
Flocculation chamber
(if required)
Influent
Sludge
Clarified
effluent
Flotation
chamber
Flocculating agent
(if required)
Downloaded by [Hamidi Aziz] at 16:04 26 September 2017
Air
FIGURE 5.5 Split flow operational modes. (From Adlan, M.N., A study of dissolved air flotation tank design
variables and separation zone performance. PhD thesis, University of Newcastle Upon Tyne, 1998.)
Sludge
Flocculation chamber
(if required)
Flotation
chamber
Influent
Clarified
effluent
Flocculating agent
(if required)
Air
FIGURE 5.6 Recycle flow operational mode. (From Adlan, M.N., A study of dissolved air flotation tank
design variables and separation zone performance. PhD thesis, University of Newcastle Upon Tyne, 1998.)
thickening mostly use circular tanks in the flotation process, which is carried out in small-size flotation plants, and require no pre-flocculation prior to flotation. However, there are large flotation plants
that use circular flotation tanks and include the flocculation process. Here, the flocculation and flotation processes are contained within the same circular tank to achieve an even distribution of bubbles
and particles/flocs attachment. In contrast, the advantages of using rectangular flotation tanks are its
simple design that makes the introduction of flocculated water into, and removal of the floated sludge
from, the tank easier; simple dimensional scale-up; and the smaller area than the circular tank.
In rectangular tanks, an inclined baffle is fixed (60° to the horizontal or at 90°) between the
contact zone and the separation zone. The baffle is fixed to elevate the bubble–floc agglomerates
toward the surface. At the same time, the baffle also reduces the turbulence condition created by
water/wastewater entering the separation zone. If the water/wastewater enters the separation zone
at high velocity, it would create a turbulence condition, which would disturb the floated sludge layer
150
Waste Treatment in the Service and Utility Industries
Contact zone influent
White
water
blanket
Separation zone
Influent
Downloaded by [Hamidi Aziz] at 16:04 26 September 2017
Contact
zone
influent
Saturator
Effluent
Effluent
recycle
Pressurized recycle
FIGURE 5.7 Rectangular flotation tank with recycle flow system. (From Murshed, M.F., Removal of turbidity, suspended solid and aluminum using DAF pilot plant. MSc thesis, Universiti Sains Malaysia, 2007.)
that accumulates continuously on the surface of the flotation tank. Figure 5.7 shows a typical setup
of a recycle flow DAF system with a rectangular flotation tank.
As can be seen in Figure 5.7, the flotation tank is divided into two zones: the front zone, which
is the contact zone or reaction zone, and the separation zone. A baffle is fixed between the contact
zone and the separation zone. The contact zone is designed to form bubble–floc agglomerates.
Small air bubbles are introduced into the contact zone. To obtain the preferred bubble size, air is
dissolved in a saturator under pressure in the range of 400–600 kPa. Thus, the pressure and the
recycle flow control the total amount of air introduced into the contact zone [26]. Once the pressurized water is released into the flotation tank at atmospheric pressure, bubbles are produced. In
DAF, small bubbles are required to achieve a good solid–liquid separation. Bubble size in the range
of 50–100 μm is the most suitable for the DAF process. If the bubbles are larger than this range,
they may create turbulence in the flotation tank and, at the same time, decrease the surface area of
the bubble–particle attachment. Once the bubbles and particles come into contact through adhesion, trapping, or absorption process in the reaction zone, the bubble–floc aggregates move to the
separation zone. Here, the bubble–floc aggregates will rise steadily to the surface of the flotation
tank, while the treated water/wastewater will be withdrawn from the bottom of the tank. Later, the
bubble–floc aggregates move to the separation zone. Here, the floc rises to the surface and floats
as a thick layer of sludge. The rising velocity of the bubble–floc aggregates can be estimated using
Stokes’ law [27]. The aggregates that do not reach the surface are swept out with the clarified water.
5.4 THEORY OF DAF
To achieve a good solid–liquid separation using DAF for influents containing particles and natural
color, coagulation or flocculation is necessary prior to the introduction of microbubbles to form
bubble–floc aggregates [19]. The main idea in DAF is to float the particles having specific gravity
more or less equal to the specific gravity of water. This should be carried out using a low-density
gas bubble, usually air. The air bubbles adhere with the particles and reduce the specific gravity to
<1.0, aggregate the particles, and float them to the surface of the flotation tank [28]. In DAF, there
are three main processes for removal of particles: bubble formation, bubble–particle attachment,
and flotation of the bubble–particle agglomerate [29].
Dissolved Air Flotation (DAF) for Wastewater Treatment
151
5.4.1 Bubble Formation
There are two processes involved in bubble formation. First, the nucleation process, which initiates
as soon as the pressurized water with air is released through the nozzle. The second step occurs when
the excess air in the saturated water is transferred to the flotation tank in the form of gas. In this step,
the bubbles begin to increase in size due to coalescence and decrease in hydrostatic pressure as they
rise through the flotation tank. However, during this step, the air volume remains constant.
5.4.2 Bubble–Particle Attachment
There are three types of bubble–particle attachment mechanisms:
Downloaded by [Hamidi Aziz] at 16:04 26 September 2017
1. Precipitation or collisions.
2. Bubbles trapped in a floc structure as the rise through the liquid medium.
3. Bubbles absorbed in a floc structure as the floc is formed.
The adhesion or collision, trapping, and absorption are sequence mechanisms in the bubble–particle
attachment (Figure 5.8a through c). The particle must be destabilized to obtain good agglomeration
between particle and bubble. However, the particle has to fulfil two most significant conditions: neutral
charge and hydrophobic surface. Under these conditions, attachment between particle and bubble is
very strong, resulting in successful flotation. Influent required coagulation process, a proper dosing
and pH value resulting in low particle charge and the formation of hydrophobic particle surface.
5.4.3 Flotation of Bubble–Particle Agglomerate
In the DAF process, the particles float due to bubbles that reduce the density of the bubble–particle
agglomerates. As long as bubble–particle agglomerates have lesser density than water (1.00 g/cm3),
they will rise and float to the surface. If the particles are small, fewer bubbles are required to
decrease the density compared with big particles that require more bubbles. The bubble–particle
agglomerates should rise to the surface of the flotation tank [30]. The agglomerates that do not reach
the surface are swept out with the clarified water. The rising velocity of the bubble–particle can be
estimated using Stokes’ law.
The three main theories in DAF show that there are some factors affecting the DAF system that
should be taken into consideration before utilizing the DAF system in water or wastewater treatment. Therefore, the system operation and all factors should be accounted for in designing and
utilizing the DAF system.
5.4.4 Kinetics of Flotation
Harper indicated that experiments seldom agree with the prediction that a bubble rising in a
Newtonian liquid can be treated in isolation, unless great care is taken to remove impurities [31].
A bubble with constant surface tension rising under gravity will rise steadily if:
1. Its motion is stable relative to random small disturbances, and
2. The time taken to approach close to terminal velocity is much less than the time required
for the bubble to change its size significantly.
At a low Reynolds number, the retarding or drag force is parallel and opposite to the terminal velocity with a magnitude of
D = 6 πaµU ,
(5.1)
152
Waste Treatment in the Service and Utility Industries
(a)
Collision of rising air bubble
and suspended solid
Precipitation of air on
the solid surface
Solid particle or solid globule
Rising air
bubble
Air bubble
formation
Downloaded by [Hamidi Aziz] at 16:04 26 September 2017
(b)
Air bubble
grows as pressure
is released
Floc structure
Air bubbles
(c)
Air bubbles
formation
Suspended solids
Rising air bubbles
Air bubbles are trapped
within the floc or on the
surface irregularites
Air bubbles
Rising air bubble
Rising air bubble
FIGURE 5.8 Bubble–particle attachment mechanism (a)–(c). (a) Adhesion of air bubble, (b) rising air
attaches to floc structure, and (c) entrapment within floc structure during formation. (From Adlan, M.N., A
study of dissolved air flotation tank design variables and separation zone performance. PhD thesis, University
of Newcastle Upon Tyne, 1998.)
where
D = drag force (kN)
a = radius of bubble (m)
μ = dynamic viscosity (kg/m/s)
U = terminal velocity (m/s)
Equation 5.1 was obtained by Stokes for slow motion of a sphere in viscous fluid [32,33]. The
expression is usually known as Stokes’ law for the resistance to a moving sphere [34]. The derivation of Stokes’ law is based on the assumption that the motion of the spherical particle is extremely
slow and that the liquid medium boundary is at an infinite distance from the particle and also is of
a large volume compared with the dimensions of the particle [35]. Clift et al. showed that bubbles
are closely approximated by spheres if the interfacial tension and/or viscous forces are much more
important than inertia forces and the term “spherical” can be used if the ratio of minor axis to major
axis lies within 10% of unity [36].
Dissolved Air Flotation (DAF) for Wastewater Treatment
153
When a solid sphere falls vertically in a liquid, the viscous liquid produces a terminal velocity
U. By equating the weight of the sphere to the sum of the up thrust and the drag [37], the following
equation is obtained:
4 3
4
π a σg = π a3ρg + 6 π aµU
3
3
U=
2
g
(σ − ρ)a 2 ,
9
µ
(5.2)
Downloaded by [Hamidi Aziz] at 16:04 26 September 2017
where
a = radius of sphere (m)
σ = density of sphere (kg/m3)
ρ = density of liquid (kg/m3)
Packham and Richards indicated that the alum sludge from a DAF water treatment plant rose at a rate of
20–35 mm/s [38]. The rising velocity is far greater than the settling velocity of an aluminum or iron floc
encountered in a water treatment works (i.e., normally <0.5 mm/s). Basing their judgment on the fact
that the rising rate of the floc in flotation was far greater than the settling rate, Packham and Richards
considered the rate of separation of suspended matter in the flotation process from the viewpoint of
Stokes’ equation governing the motion of a sphere through a viscous medium and thus showed that
Equation 5.2 was appropriate to describe the rising rate of the particle in the flotation process. Packham
and Richards, in reviewing Equation 5.2, were of the opinion that if the size of the suspended matter is
increased, a higher separation rate may be achieved. This is because the rate of separation is directly
proportional to the square of the radius of the particles, the difference in the densities of liquid and the
suspended particles, and inversely proportional to the liquid viscosity, as shown by Equation 5.2.
Research carried out in Russia [39] showed that at small Reynolds numbers, gas bubbles moved
like solid spheres. Theoretical values of bubble rise velocity in water were not in agreement with
much experimental data. For a gas bubble, which is assumed to behave like a solid, the surface can
sustain a finite shear stress, the tangential velocity of the surface is everywhere zero relative to the
center of the bubble, and the conventional Stokes’ solution applies. According to Jameson, a force
balance equation will result [40]:
6 πµUa =
4 3
πa (ρ − ρg ) g.
3
(5.3)
When the density of gas ρg is negligible compared with the density of liquid ρ, the terminal velocity is given by
U=
2ρga 2
.
9µ
(5.4)
Equation 5.4 shows that the rise velocity of a bubble is controlled by the size of the bubble and the viscosity of the fluid. If the radius of the bubble is increased, the rise velocity will be increased. The kinematic viscosity is affected by the density and the temperature of the fluid. An increase in temperature
will result in the decrease in viscosity, and hence an increase in rising velocity of the bubble. Shannon
and Buisson indicated that bubble rise rates at 80°C increased three times compared with those at
20°C [41]. Force balance is presented in terms of drag coefficient CD by Harper [31] as follows:
CD =
Force on bubble 4/3πρga 3 4 gd
=
=
.
1/2ρU 2 πa 2
1/2ρU 2 a 2 3U 2
(5.5)
154
Waste Treatment in the Service and Utility Industries
This coefficient is the force per unit cross-sectional area, made dimensionless by the dynamic
pressure 1/2ρU 2. Substituting Equation 5.4 in 5.5 yields:
Downloaded by [Hamidi Aziz] at 16:04 26 September 2017
CD =
24
,
Re
(5.6)
where Re is the Reynolds number. Equation 5.6 is used for viscous resistance at low Reynolds numbers, Re < 0.5 [42]. The same equation was used by Vrablik to determine the maximum bubble size
of 130 μm for a complete viscous flow [43]. He indicated that the maximum value of the Reynolds
number for laminar or viscous flow is 1.13. The relationship governing bubble size, laminar flow,
bubble rising velocity (as per Equation 5.4), and temperature has been established. This relationship
is shown in Table 5.1. The optimum bubble rising velocity for the DAF system is about 300 mm/min
[44]. The rising velocity should not be <125 mm/min or >500 mm/min.
Experimental work by Fukushi et al. showed that Equation 5.4 cannot be used to describe bubble
rise velocity [44]. This is due to the turbulent environment that occurs in the mixing zone (reaction
zone) of the DAF tank. They suggested that the following equation is more appropriate and that it
agreed with their experimental results:
U=
pga 2
.
3µ
(5.7)
The properties of air bubbles produced in the DAF process developed by Fukushi et al. were compared with those developed by Edzwald et al. [45,46]. There are many differences between the
models. These are shown in Table 5.2. The model developed in 1985 by Fukushi et al. was based
TABLE 5.1
Relationship between Bubble Size, Rise Velocity, Temperature, and Laminar Flow
Bubble Size
(μm)
10
20
30
40
50
80
110
120
130
140
160
170
Rise Velocity (m/h) Above
Which Turbulent Flow Existsa
Terminal Rise Velocity
(m/h) Based on Stokes’ Law
4°C
20°C
4°C
20°C
565
283
188
141
113
70.7
51.4
47.1
43.5
40.4
35.3
33.2
360
180
120
90
72
45
32.7
30
27.7
25.7
22.5
21.2
0.125
0.499
1.12
2.00
3.12
7.99
15.1
18.0
21.1
24.5
31.9
36.1b
0.196
0.783
1.76
3.13
4.89
12.5
23.7
28.2
33.1b
38.3b
50.1b
56.5b
Source: Malley, J.P. Jr., A fundamental study of dissolved air flotation for treatment of low turbidity waters
containing natural organic matter. Unpublished Ph.D. dissertation, University of Massachusetts,
1988.
a Based on a critical Reynold’s number of 1.0 for the upper limit of laminar flow.
b Indicates that the terminal rise velocity will result in turbulent flow.
155
Dissolved Air Flotation (DAF) for Wastewater Treatment
TABLE 5.2
Models Developed for the Dissolved Air Flotation Process
Fukushi et al. [47]
Downloaded by [Hamidi Aziz] at 16:04 26 September 2017
Size range da (μm)
Rise velocity (cm/s)
Zeta potential (mV)
Pressure P (kPa)
Recycle ratio r
Concentration na (cm−3)
Size range df (μm)
Density ρf (g/cm3)
Suitable mobility (μm/s/V/cm)
Collision model
Flow regime
Mechanism
Attachment mechanism
Rise velocity of agglomerate (cm/s)
Edzwald et al. [46]
Generated Air Bubbles
10–120 (average 60)
10–100 (average 40)
gda2/12ν
−150 at pH 7
392
0.1
104–105
gda2/18ν
Not measured
345–585
0.08
104–105
Produced Flocs
100–103
Floc density function
0 to +1 (clay floc)
−1 to +1 (colour floc)
Bubble–Floc Collision and Attachment
Population balance model
Turbulent flow
Locally isotropic turbulence, viscous
subrange diffusion
Electrical charge interactions (coverage of
precipitated coagulant on a floc surface)
0.1–2.6 (observed)
100–102 (10–30 μm is best)
1.01 (assumed)
0.5 or less
Single collector collision model
Laminar flow
Brownian diffusion,
interception, gravity settling
Electrical charge interaction,
water layer at floc surface
About 0.3 (nearly equal to
bubble rise velocity)
Source: Fukushi, K., et al., Water Sci. Technol., 31, 37–47, 1995.
g, gravity; da, diameter of bubble; ν, kinematic viscosity; df, diameter of floc; ρf, density of floc.
on the population balance model of bubbles and flocs in a turbulent flow environment Population
Balance Model (PBM model) [47]. However, the model developed by Edzwald was derived from
a single collision theory in a laminar flow condition Single Collision Theory (SCT model). In the
SCC model, collision occurs due to Brownian diffusion, interception, and gravity settling. Fukushi
et al. indicated that Brownian diffusion and gravity settling cannot be dominant for a normal floc
(10–1000 μm) and bubble size range in flotation [45]. Interception also cannot be dominant because,
in practice, the mixing zone is apparently in a turbulent flow where certain energy dissipation occurs.
In fact, a literature survey indicates that Equation 5.7 was initially suggested by V. G. Levich in
1962. Levich showed that Equation 5.7 is applicable for small Reynolds numbers, Re << 1 and when
the following inequality holds [48]:
ga 3
<< 1,
3ν2
(5.8)
where υ = µ ρ (i.e., kinematic viscosity equals dynamic viscosity divided by density of liquid).
If the medium is water, the size of moving bubbles will be a << 2 × 10−2 cm. Levich also indicated that the theoretical value of the drag coefficient for a gas bubble in water is equal to 8/Re
(i.e., one and one-half times smaller than for a solid sphere). This value is not in agreement with
Equation 5.6. However, Levich indicated that Allen’s experimental results with small Reynolds
156
Waste Treatment in the Service and Utility Industries
numbers completely disagreed with the theory and led to values for the drag coefficient that coincided exactly with the drag on a solid sphere [48].
A mathematical equation for solid/liquid separation was developed by Howe limited to flotation
of discrete particles without the interference of surface-active forming agents. It was derived from
a differential equation of motion, which was expanded to give solutions for the rising velocity of
a particle with changes in the applied rising force, particle diameter, liquid viscosity, and particle
density [49]. The equation is as follows:
Downloaded by [Hamidi Aziz] at 16:04 26 September 2017
R = 1− e
V
− r
Q Ah
,
(5.9)
where
R = ratio of total removal of solid concentration after flotation to the inflow solid concentration
= 1 − Co Ci
Co = the effluent suspended solids
Ci = the influent suspended solids
Vr = the rising velocity of a single particle/air bubble (m/s)
Q = the flow applied to the flotation unit (m3/s)
Ah = the horizontal area of the unit (m2)
Equation 5.9 is limited to discrete particles without the interference of surface-active forming
agents. Karamanev, in his article on the rise of bubbles in quiescent liquid, showed that equations based on the model of bubble with internal circulation often fail to describe the real systems
adequately [50]. This is because even highly purified liquids (such as triple distilled water) contain
enough surface-active components to affect internal bubble recirculation. Recirculation is normally
due to the presence of surface-active substances, and the resulting variable surface tension leads to
a change in boundary conditions of the bubble [48]:
U = 25V 1/6 ,
(5.10)
where V is the volume of the bubble. However, this equation works only for large, spherical capshaped bubbles. The drag coefficient CD of the gas bubble calculated on the basis of equivalent
sphere diameter by most authors was found to have a large deviation of CD as a function of the
Reynolds number when different liquids are used. The assumption made by most authors that freefalling heavy spheres behave exactly like free-rising solid spheres is found to be incorrect, especially for particles with densities <0.3 gm/cm3 and Re > 130 rising in water. Karamanev suggested
the following equation based on the balance of forces acting on a rising bubble [50]:
1
CD SρU 2 = ∆ρgV ,
2
(5.11)
where ρ is the liquid density, ∆ρ is the difference of density between liquid and gas, and S is the
area of the bubble. To obtain CD based on real bubble geometry, the area S should be determined
from the diameter projected on the horizontal plane circle, dh; S = πdh2 4. Then, the volume of
the bubble is calculated using the equivalent diameter, V = πde3 6. These values are substituted in
Equation 5.11 and become
CD =
4 g∆ρde3
.
3ρdh2U 2
(5.12)
Dissolved Air Flotation (DAF) for Wastewater Treatment
157
Equation 5.11 can be written in terms of U:
12
8 gV
U =
.
πCD dh2
(5.13)
By substituting from Equation 5.12,
12
8g
d
U = 2 3 1 3 V1 6 e .
π
C
6
d
D
h
(5.14)
For Re < 130, Karamanev suggested that CD can be calculated using the following equation:
Downloaded by [Hamidi Aziz] at 16:04 26 September 2017
CD =
(
24 1 + 0.173Re 0.657
Re
)+
0.413
.
1 + 16300 Re−1.09
(5.15)
For spherical bubbles at Re < 1, CD = 24 Re and de dh = 1 = 1 and Equation 5.12 transforms to Stokes’
equation.
5.4.5 Solubility of Air
In flotation, the quantities of air used are normally expressed in terms of volume of air supplied per
volume of water treated [10]. Henry’s law is used when treating saturated water as a dilute solution
of air in water [43]. It must be remembered that Henry’s law was originally based on his experiment
with N2, O2, N2O, H2S, CO2 and water at only one temperature. The concept that the law could be
used for general application is unfounded [51]. However, experimental work on wastewater with
dissolved solids up to 1000 mg/L with pressures up to 500 kPa showed that Henry’s law constant
could be used to calculate the mass of dissolved air [52]. For ideal dilute solutions where the solute
obeys Henry’s law but not Raoult’s law, and the solvent obeys Raoult’s law, then the use of Henry’s
law is applicable [53,54].
PB = xB K B ,
(5.16)
where PB is the vapour pressure, x B is the mole fraction of the solute, and KB is constant. Based on
Henry’s law, Edzwald and Walsh suggested the following:
CS = f
p
,
k
(5.17)
where CS is the concentration of air in the saturated liquid, p is the absolute pressure, k is the
Henry’s law constant, and f is the efficiency factor, which is about 70% for unpacked saturators and
up to 90% for packed systems. Values of k at 0°C and 25°C are 2.72 and 4.53 kPa/mg/L, respectively. However, others indicated that Henry’s law is not strictly applicable when treating saturated
water [55,56]. The equation has to be modified with an exponent m on the pressure p as follows [56]:
CS =
pm
.
k
(5.18)
Klassen and Mokrousov, in their review on the solubility of gases in water, were of the opinion
that the solubility of gases depends on the partial pressure, temperature, and concentration of other
substances in the solution [57]. If the partial pressure is increased, then the solubility of gas will be
158
Waste Treatment in the Service and Utility Industries
increased. However, if the concentration of soluble substances in water is increased, gas solubility
will be decreased as definite quantities of water molecules complex in the form of hydrated ions.
Edwards added that if the total pressure is <507 kPa (5 atm.), the solubility for a particular partial
pressure of solute gas is normally independent of the total pressure of the system [58]. In its relationship to temperature, the solubility of a gas will be decreased when the temperature is increased
[43,59]. This is as illustrated in Figure 5.9. In the case of distilled water, when the temperature is
increased from 0°C to 30°C, the solubility of air is reduced by 45%. Liquid solubility of the gases
varies as shown in Table 5.3.
180
160
Air dissolved (mg/L)
Downloaded by [Hamidi Aziz] at 16:04 26 September 2017
140
120
100
80
60
T= 5°C
40
T=10°C
T=15°C
20
0
T=20°C
0
100
200
Pressures kPa
300
400
500
FIGURE 5.9 Solubility of air in water. (From Adlan, M.N., A study of dissolved air flotation tank design
variables and separation zone performance. PhD thesis, University of Newcastle Upon Tyne, 1998.)
TABLE 5.3
Solubility of Various Gases at 20°C and 760 mm Hg
Type of Gas
Nitrogen
Oxygen
Hydrogen
Carbon dioxide
Carbon monoxide
Air
Hydrogen sulfide
Sulfur dioxide
Cubic cm Gas/Cubic cm Water
Gram of Gas/100 gm of Water
0.015
0.031
0.018
0.88
0.023
—
2.58
39.4
0.0019
0.0043
0.00016
0.17
0.0028
1.87
0.38
11.28
Source: Vrablik, E.R., Proceedings of the 14th Industrial Waste Conference, Purdue University, West
Lafayette, IN, 743–779, 1959.
Dissolved Air Flotation (DAF) for Wastewater Treatment
159
Bratby and Marais in their studies on saturator performance indicated that it would be difficult
to achieve full saturation at a saturator pressure <350 kPa [60]. From an economic point of view, the
efficiency of the saturator system was important. They found that by using a packed system of 0.5 m
depth with Raschig rings of 25 mm diameter, full saturation was achieved at saturator pressures
beyond 250 kPa for a surface loading up to 2500 m/day. A similar level of saturation was found by
Zabel and Hyde using a packed saturator of 0.75 m depth with 25 mm Berl saddles [23].
For design purposes, Bratby and Marais suggested that at a temperature of 20°C with a pressure
of 3 atm., the concentration of air precipitated on reducing the pressure to atmospheric pressure is
given by the following equation [60]:
aP = 19.5 p (mg L ),
(5.19)
Downloaded by [Hamidi Aziz] at 16:04 26 September 2017
where p is the saturator pressure in atmospheres.
5.4.6 Bubble Generation
Rykaart and Haarhoff indicated that the geometrical design and operating conditions of the injection nozzles were important determining factors for bubble size [61]. They reported that saturator
pressure does not have a consistent effect on nozzle efficiency. There were contradicting claims
regarding whether a higher pressure produces smaller bubbles [55,62] or bigger bubbles [52,63]. But
Jones and Hall reported that there was no significant relationship between pressure variation and
bubble size [64].
Studies by Bratby and Marais [65] showed that the shape and roughness of the valve, the degree of
turbulence and dilution of saturator feed downstream of the valve, and the concentration of particulate nuclei in the dilution water had a negligible effect on the precipitation of air from a solution (i.e.,
mass of air precipitated to unit volume of saturator feed). However, these findings were contradicted
by those reported by others [55,61] in terms of the shapes and roughness of the valves. Rykaart and
Haarhoff showed that at a saturator pressure of 500 kPa, a nozzle with a bend in its channel produced
a bubble size of 49.4 μm (median diameter) compared with a nozzle with a tapering outlet that produced 29.5 μm [61]. When the saturator pressure was reduced, the bubble sizes were reduced.
For a continuous-flow DAF plant, Edzwald and Walsh predicted that the concentration of air
released in the tank (Cr) would be as follows:
CS − Ca
Cr =
Rr − K ,
1 + Rr
(5.20)
where Ca is the concentration of air that remains in solution at atmospheric pressure, Rr is the recycle
ratio, which is equal to the recycle flow rate divided by the influent flow rate, and K is the influent
saturation factor defined as (CS − Ca ), where Co is the concentration of air in the influent water. In
most cases, Co is saturated, and this means K = 0. To find the bubble volume concentration (ϕb),
Edzwald and Walsh suggested that Cr be divided by the saturated density of air ρsat as shown in the
following equation,
φ b = Cr ρsat .
(5.21)
To get the generated air volume at the same temperature under atmospheric pressure, Takahashi
et al. used the following equation by considering air as an ideal gas [55]:
ρ P − P RT
VA = w A O
,
M w PO H E
(5.22)
160
Waste Treatment in the Service and Utility Industries
where ρw is the density of water in gm/cm3, PA is the dissolved pressure in dyne/cm2, PO is the atmospheric pressure in dyne/cm2, R is the gas constant in erg k ⋅ mol, T is the absolute temperature in
Kelvin, Mw is the molecular weight of water in g/g·mol, and HE is Henry’s law constant in dyne/cm2.
By assuming that all the dissolved air in water changes into bubbles, the theoretical generated
flow rate can be obtained using Equation 5.18. The experimental results by Takahashi et al. showed
that the generated air flow rate increased with an increase in dissolved pressure and also with an
increase in liquid flow rate [55]. To obtain the volume of air occupied by a single spherical bubble,
Vb, Takahashi et al. suggested the following equation:
Downloaded by [Hamidi Aziz] at 16:04 26 September 2017
P + ρw gh + 4 σ dav π 3
Vb = O
dav ,
PO
6
(5.23)
where h is the depth from the liquid surface to the bubble in cm, σ is the surface tension in dyne/cm2,
and dav is the volumetric mean diameter of the bubble in cm. According to the authors, the effect of
liquid depth is negligible; thus, the measurement of the bubble diameter was carried out at the top
of the flotation tank. The number of bubbles generated per cm3 of water could be obtained from the
following equation:
Nb =
GQ
,
Vb
(5.24)
where G is the volumetric flow rate of air generated under decrease in pressure (cm3/s), and Q is the
volumetric flow rate of liquid (cm3/s). Their experimental results showed that by increasing the dissolved pressure and liquid flow rate, the number of bubbles will be increased. The geometry of the
nozzle also affects the bubble size. By using a needle valve, Takahashi et al. obtained the following
equation:
2
P −P
N b = 1 × 10 4 A O Q.
PO
(5.25)
The calculated and experimental values of the number of bubbles were compared, and the results
were claimed to be remarkably in agreement with the equation used.
For an efficient solid–liquid separation process, small bubbles are needed [66–68]. Bubble sizes
in the range of 20–80 μm are capable of good attachment to floc particles. Larger bubbles will create a hydraulic disturbance along their rising path toward the surface and a decrease in the surface
area. For example, a 2 mm bubble contains the same amount of air as 64,000 bubbles of 50 μm in
size. Collin and Jameson reported that the optimum bubble size in the microflotation process is
approximately 50 μm [67].
5.4.7 Collision
Reay and Ratcliff, in their study of dispersed air flotation, defined the collection efficiency of a
bubble as the fraction of particles in the bubble’s path that are picked up by the bubble [69]. Particles
of about 3 μm diameter or larger will not be affected by Brownian motion. They will be in contact
with the bubble only if their hydrodynamically determined trajectories come within one particle
radius (rp) of the bubble. This region is called the collision regime. By considering the collision
regime in which the Brownian diffusion is negligible [70], the collection efficiency of a bubble can
be expressed as:
η = η1 × η2 ,
(5.26)
Dissolved Air Flotation (DAF) for Wastewater Treatment
161
where
ƞ1 = collision efficiency, that is, the fraction of particles in the bubble’s path that collided with
the bubble
ƞ2 = attachment efficiency, that is, the fraction of particles colliding with the bubble that
stick to it
Downloaded by [Hamidi Aziz] at 16:04 26 September 2017
Equation 5.26 indicates that ƞ2 will depend mainly on the chemical nature of the particle surface,
the bubble surface, and the thin film of liquid draining from between them. Reay and Ratcliff also
1.9
reported on the predicted collision efficiency and illustrated it with a graph; η1 = 1.25 ( rp Rb ) for
2.05
2
( ρp ρf ) = 1, η1 = 3.6 ( rp Rb ) for ( ρp ρf ) = 2.5, and ƞ1 is roughly proportional to ( rp Rb ) over the
density range used. The symbols used in the aforementioned expressions are as follows [69]:
rp = particle radius (cm)
Rb = bubble radius (cm)
ρp = particle density (gm/mL)
ρf = fluid density (gm/mL)
Since ƞ1 is proportional to Rb2, the average number of particles picked up by a bubble (by assuming ƞ1 is constant) should be roughly independent of bubble size and the flotation rate should
be proportional to bubble frequency (i.e., the amount of bubbles rather than bubble diameter
over the entire range of particle sizes). This prediction is applicable to bubbles of diameter up
to 0.1 mm [71]. However, when latex particles (3–9 μm) having almost the same density as water
and larger zeta potential (+10.6 mV) were used in the experiments, they could not get close to
the bubble surface [71]. This means the bubble–particle collision model is not appropriate for
latex particles.
Flint and Howarth, in their review on the collision efficiency of small particles with spherical
air bubbles, reported that the collision of a particle with a bubble would depend on the balance of
viscous, inertial, and gravitational forces acting on the bubble [72]. Besides that, the form of streamlines around the bubble also play an important role in whether or not collision takes place. Flint and
Howarth formulated an equation of motion of a small spherical particle similar to a spherical bubble
rising in an infinite pool of liquid in the jth direction as follows [72]:
mp
av j
= G j + Cd (u j − v j ) ,
at
(5.27)
where
Gj = body force acting on the particle; for raindrop collision, Gj = 0. In flotation, there is clearly
a component of relative acceleration due to gravity because the bubble and particle are of
distinctly different densities.
CD =dimensional drag coefficient for the particle, depending on the shape of the particle and the
Reynolds number past it. For a spherical particle, the drag will be the same in all directions.
Vj = particle velocity.
uj = velocity the fluid would have at the position of the particle if no particle were there. For fine
particles in flotation, it is assumed that the flow around the particle has an insignificant effect
compared with the flow pattern due to the bubble; uj then depends on the shape of the bubble
and the Reynolds number around it.
4t = time
By considering the relative two-dimensional motion of a spherical bubble and particle where the
bubble is held stationary at the origin of the coordinate system by a liquid flow equal to the bubble
162
Waste Treatment in the Service and Utility Industries
y
U
Vc
Vr
G
r
Bubble
Particle
rb
Downloaded by [Hamidi Aziz] at 16:04 26 September 2017
C
x
FIGURE 5.10 Geometry of bubble–particle system. (From Adlan, M.N., A study of dissolved air flotation
tank design variables and separation zone performance. PhD thesis, University of Newcastle Upon Tyne, 1998.)
rise velocity in the negative direction (Figure 5.10), Flint and Howarth suggested the equation of
motion for the particle as follows [72]:
4 3 ∂vy
πrp ρp
= 6 πµ f rp (u y − vy )
∂t
3
(5.28)
4 3 ∂vx 4 3
πrp ρp
= 3 πrp (ρp − ρf ) g − 6 πrp (u x − vx ) .
∂t
3
(5.29)
Reducing these equations to their dimensionless form and introducing the variable v, u, and t, and
parameters K and G:
v*x =
vx * vy
vy =
u
u
ux * uy
uy =
u
u
u *x =
t* =
tu
rb
and,
K=
2ρp rE2 u
rb
9µ f
G = 2 (ρp − ρf ) rp2
g
u
9µ f
i.e.,
K
K
∂v*y
= u *y − v*y
∂t *
∂v*x
= −G − u *x + v*x
∂t *
Dissolved Air Flotation (DAF) for Wastewater Treatment
163
Downloaded by [Hamidi Aziz] at 16:04 26 September 2017
where
rp = particle radius (cm)
ρp = particle density (gm/mL)
ρf = fluid density (gm/mL)
v = component of particle velocity (cm/s)
t = time (s)
μf = fluid viscosity (kg/m/s)
u = component of velocity field due to bubble (cm/s)
x, y = Cartesian position coordinates (cm)
v = dimensionless component of particle velocity (cm/s)
t = dimensionless time (s)
K = particle inertia parameter
G = dimensionless settling velocity of particle (cm/s)
According to Flint and Howarth, calculation for K down to 0.001 shows that the collision efficiency
remains substantially constant for 0.001 < K < 0.1, meaning that collision efficiency is virtually independent of K and of whether Stokes or potential flow is assumed [72]. They suggested that for a fine
particle characterized by K < 0.1, inertial effects of the particle may be neglected, and single bubble
collision efficiency η can be calculated from:
η=
G
.
(1 + G )
(5.30)
However, Flint and Howarth indicated that in the flotation tank, the collision efficiency may be
several times as great as those predicted from single bubble calculations [72]. This may be due to
at least three reasons:
1. The presence of hindering effects of the neighboring bubbles that reduced the rising velocity of the bubble. For fine bubbles, this could lead to an increase in collision efficiency.
2. Difference in the shape of liquid streamlines around the bubble. The greater the number
of bubbles, the closer the assemblage and straighter the streamlines. This results in the
increase of collisions between particles and bubbles.
3. The motion of particles upstream from the target bubble is influenced by the layers of
bubbles ahead and is thus no longer parallel to the direction of bubble motion.
The aforementioned opinion, which was expressed by Flint and Howarth, is found to be in agreement with that of Fukushi et al., as the latter showed that a single collector collision model was not
appropriate in the DAF process [45]. Furthermore, King indicated that the calculated collision efficiency based on the works of Reay and Ratcliff, Flint and Howarth, and Sutherland and Woodburn
et al. were not in agreement with each other [73,71–72,74–75].
5.4.8 Interception and Diffusion
According to Yao et al., a single particle of filter media is a collector, and if any suspended particle
is in contact with the collector, then a process known as interception occurs [75]. The contact efficiency of a single media particle or collector is the ratio of the rate at which the particles strike the
collector to the rate at which particles flow toward the collector, which can be expressed as follows:
η=
Rate at which particles strike the collector
,
πd 2
uo co
4
(5.31)
164
Waste Treatment in the Service and Utility Industries
where
uo = water velocity (m/s)
co = suspended particle concentration upstream from the collector where the flow is undisturbed
by the presence of the grain
d = grain diameter (cm)
In the case of flotation, the single collector efficiency (ƞ) may be defined as follows [46,77]:
Downloaded by [Hamidi Aziz] at 16:04 26 September 2017
η=
Particle − buble collision rate
.
Particle − bubble approach
(5.32)
Reay and Ratcliff indicated that submicron particles will reach the bubbles mainly by Brownian
diffusion. In the diffusion regime, collection efficiency will decrease with increasing particle
radius, rp [69]. Flotation of these submicron particles could be improved if they were agglomerated
into flocs of suitable size in the collision regime. Theoretical calculations were made on particles
with diameter < 0.2 μm and bubbles size of 75 μm. At normal temperatures and pressures, particles
< 1 μm in diameter suspended in gases or water will exhibit a Brownian motion that is sufficiently
intense to produce collision with a surface immersed in the fluid [78]. Yao et al., in describing basic
transport mechanisms in water filtration, explained that when a particle in suspension is subjected
to random bombardment by molecules of the suspending medium, then a Brownian movement of
the particle known as diffusion takes place [76]. Numerical and analytical determinations of singlecollector efficiency were discussed by Yao et al. based on the works of previous investigators, and
the following equations were established:
ηD = 4.04Pe
−2 3
kT
= 0.9
µdp dvo
d
ηI = 23 p
d
ηg =
(5.33)
2
(5.34)
(ρp − ρ) gd 2 ,
18µvo
2 3
p
(5.35)
where ƞD, ƞI, and ƞG are the theoretical values for single-collector efficiency when the sole transport
mechanisms are diffusion, interception, and gravity settling, respectively. Pe is the Peclet number
(i.e., Pe = 2 RbU b Df , where Rb is bubble radius, Ub is bubble rising velocity, and Df is particle diffusivity in cm2/s), k is Boltzmann’s constant, T is the absolute temperature, dp is the diameter of
suspended particle, d is the diameter of the collector or the bubble, which is equal to dp, vo is the
approach velocity of fluid, and ρ is the density of fluid, which is equal to ρf.
Then the expression for total single-collector efficiency of a media grain can be written as follows [76,79]:
ηT = ηD + ηI + ηG .
(5.36)
Edzwald and Walsh used the same theoretical approach as in filtration to develop a conceptual
model for flotation [10,76,79]. Thus, the following equations are introduced:
kbT
ηD = 0.9
µdp dbU b
2 3
(5.37)
Dissolved Air Flotation (DAF) for Wastewater Treatment
d
ηI = p
db
2
3
2
ηG = (ρp − ρf )
165
(5.38)
gdp2
.
18µU b
(5.39)
Comparing the equations used in filtration to the aforementioned equations, the approach velocity of
fluid and Boltzmann’s constant have been changed to Ub (bubble rise velocity) and k b (Boltzmann’s
constant for bubble), respectively. This is done to suit the mechanisms involved in flotation. Ward,
in his review on capture mechanisms, introduced a new form of equations for ηD and ηG by substituting the bubble rise velocity U from Equation 5.4 into Equations 5.37 and 5.39. Thus, we have the
following result [56]:
Downloaded by [Hamidi Aziz] at 16:04 26 September 2017
2 3
kT 1 2
ηD = 6.18
ρf gdp db
(5.40)
d (ρp − ρf )
ηG = p
.
ρf
db
(5.41)
2
Results on single-collector efficiency by Edzwald and Walsh show a minimum efficiency occurs at
a particle size of around 1 μm [10]. For removal efficiency, Edzwald et al. used the same principle as
in Equation 5.26, changing only the symbols of the expression as follows:
R = α E ηT (100%),
(5.42)
where αE is the attachment efficiency. If the total number of bubbles (Nb) is considered, then Edzwald
et al. suggests the following equation for particle removal:
dN p
= − (α pb ηT ) AbU b N b N p ,
dt
(5.43)
where Ab is the projected area of the bubble and Np is the particle number concentration. By having
a bubble volume concentration of Φ b = πdb3 N b 6, and substituting into Equation 5.43, we have the
following equation:
3 α η U Φ N
dN p
πd 2
6Φ
= − (α pb ηT ) b U b N p 3b = − pb T b b p .
2
dt
4
db
db
(5.44)
The particle number concentration removal in terms of flotation tank depth can be rewritten as:
dN b
3 α η Φ N
= − pb T b p .
dH
2
db
(5.45)
Edzwald et al. also produced a summarized table of their model parameters for DAF facilities. This
is as shown in Table 5.4. However, this model has not been tested or verified [10].
166
Waste Treatment in the Service and Utility Industries
TABLE 5.4
Model Parameters for DAF Facilities
Downloaded by [Hamidi Aziz] at 16:04 26 September 2017
Parameter
Affected by
Comments
α pb (particle-bubble
attachment efficiency)
Np (particle number
concentration)
Pre Treatment
Particle-bubble charge interaction
and hydrophilic nature of particles
Coagulation addition and
flocculation time
ηT (single collector efficiency)
Diffusion and interception
dp (bubble diameter)
Flotation Tank
Saturator pressure
Φb (bubble volume
concentration
Saturator pressure and recycle ratio
Improve α pb by chemical pretreatment,
coagulation, and pH conditions
Coagulant may add particles,
flocculation may reduce Np and
increase dp.
Minimum for dp of 1 μm
Small bubbles produce large interfacial
areas and surface forces between
bubbles and particles. Small bubbles,
improve ηT
Large Φb ensures collision opportunities
and lowering of floc density
Source: Edzwald, J.K., Walsh, J.P., Dissolved air flotation: Laboratory and pilot plant investigation. AWWA Research
Foundation and AWWA, 1992.
Ward, in his article on DAF, made an improvement on Equation 5.45 by integrating it over the
tank depth H from N = No at the surface H = 0 to N = n at the tank base H = H. Thus, the overall
particle removal equation becomes [56]:
N = N oe
3α pb ηT Φb H
−
2 db
.
(5.46)
Then, the overall efficiency is given by:
η = 1−
N
.
No
(5.47)
5.4.9 Tank Design
The usual design procedure for any flotation unit can be based on Figure 5.11. All the suspended
solids in the flotation chamber should have a sufficient rise velocity to travel the effective depth
D within the specified detention time T. This means the rise rate VT must be at least equal to the
effective depth D divided by the detention time T, or equal to the flow divided by the surface area:
VT =
D Q
=
,
T AS
where
VT = vertical rise rate of suspended solids (m/s)
D = effective depth of flotation chamber (m)
T = detention time (s)
Q = influent flow rate, (m3/s)
AS = surface area of flotation chamber (m2)
(5.48)
Dissolved Air Flotation (DAF) for Wastewater Treatment
167
VT
D
VH
L
Downloaded by [Hamidi Aziz] at 16:04 26 September 2017
FIGURE 5.11 Basic design concept of flotation unit. (From Adlan, M.N., A study of dissolved air
flotation tank design variables and separation zone performance. PhD thesis, University of Newcastle
Upon Tyne, 1998.)
The particles to be removed must also have a horizontal velocity
VH =
Q
,
AC
(5.49)
where
VH = horizontal velocity (m/s)
AC = cross-sectional area of flotation chamber (m²)
If the flotation chamber is in a rectangular shape, then the following equations can be established:
AC
D
(5.50)
AS
AS
A
=
= VH D S ,
W ( AC D )
Q
(5.51)
W=
L=
where
W is the width of the flotation chamber (m),
L is the effective length of the flotation chamber (m),
Q is the influent flow rate (m3/s),
and the value of D/W is usually between 0.3 and 0.5.
The size of the flotation tank can be reduced if the separation rate is increased [80]. Katz and
Wullschleger showed that a particle with a bubble attached to it would increase in its rising rate with
an increase in the particle size. This finding is similar to that reported by Packham and Richards [38].
However, other factors such as pressure, recycle ratio, temperature, pH, zeta potential of the particles,
number and size of bubbles produced, types of nozzles, flocculation process, flow condition, and configuration of the tank are believed to have a significant effect on the separation process [11,45,81].
Longhurst and Graham reported that the surface overflow rate (SOR) or rise rate is the fundamental criterion for tank design [15]. It is defined as the flow rate divided by the surface area of
the flotation tank. In practice, the surface area is based on the interfacial area between clarified
water and sludge, and not on the total area of the flotation tank [15]. The characteristics of water
and bubble size will determine the air/floc aggregate rise velocity. For normal design purposes,
rise velocities between 3 and 8 m/h have been used [68]. For laminar flow, the maximum size of
bubbles is 130 μm; for bubbles measuring <130 μm, Stokes’ law applies [11], and Equation 5.2 can be
168
Waste Treatment in the Service and Utility Industries
Downloaded by [Hamidi Aziz] at 16:04 26 September 2017
used to calculate the rise rate. The maximum bubble size for laminar flow can be calculated using
Equation 5.3 by assuming limited laminar flow, Re = 1, and using the relationship between bubble
size and rise rate of air bubble, which has been established in graphical form [11]. A survey carried
by Longhurst and Graham showed that the average normal operating SOR is below 6–9 m/h with a
maximum rate of up to 11 m/h [15].
Bratby and Marais, in their investigation on the application of DAF in activated sludge, were of
the same opinion as Longhurst and Graham regarding the design of flotation units [15,82]. Instead
of SOR, Bratby and Marais used the term down flow rate, which is defined as the total flow into the
unit divided by the plan area at the outlet. It is the value of limiting down flow rate (VL), where the
bubble–particle agglomerates are carried down with the effluent that controls the design of the tank.
Data published by Edzwald on the design and operation parameters of DAF showed that there were
still considerable variations in retention time, hydraulic loading, and recycle ratio between different
treatment works in different parts of the world [83]. These are shown in Table 5.5.
In terms of shape, Zabel and Melbourne indicated that a rectangular shape has gained greater
acceptance due to advantages such as simple design, easy introduction of flocculated water, easy
TABLE 5.5
Summary of DAF Design and Operation Parameters
Parameter
South Africa
Finland
The Netherlands
UK
UK
(Edzwald)
Scandinavia
20–29
18–20
28–44
5–12
8.4–10
6.7–7
Flocculation
Intensity
Time (min)
4–15
20–127
8–16
Flotation
Reaction zone
Time (min)
1–4
0.9–2.1
Hyd. load. (m/h)
40–100
50–100
Separation zone
Hyd. load. (m/h)
5–11
2.5–8
Total flotation area
Hyd. load. (m/h)
9–26
10–20
Time (min.)
Recycle (%)
11–18
6–10
5.6–42
6.5–15
6–10
5–10
10
400–550
460–550
Unpacked Sat.
Pressure (kPa)
400–600
Hyd. load. (m/h)
20–60
Time (s)
20–60
Packed Sat.
Pressure (kPa)
300–600
Hyd. load. (m/h)
50–80
Packing depth (m)
0.8–1.2
Pressure (kPa)
400–500
300–750
Saturatorsc
400–800
Source: Edzwald, J.K., Water Sci. Technol., 31, 1–23, 1995.
Unspecified with respect to unpacked or packed saturator, [1,2,15,83].
c
310–830
480–550
169
Downloaded by [Hamidi Aziz] at 16:04 26 September 2017
Dissolved Air Flotation (DAF) for Wastewater Treatment
float removal, small area, and flexibility of scale-up [21]. In addition to that, floc breakup is minimized, hydraulic efficiency is maximized, and engineering and construction is simplified [15]. A
tank with SOR in the range 6–12 m/h would have a depth of 1.2–1.6 m and a residence time of
5–15 min [84]. Results from a survey (questionnaire) carried out by Longhurst and Graham in Great
Britain showed that, in practice, tank depths range 1–3.2 m with a mean value of 2.4 m, while tank
shapes vary from “squarish” to “long and thin”; however, there is a continuing debate in this area
[15]. Gregory and Zabel indicated that tank depth of about 1.5 m with an overflow rate of 8–12 m/h
(depending on the type of water) is normally used [11]. An effective flotation unit could be between
0.48 and 2.74 m deep [6]. The angle for the inlet baffle is approximately 60° to the horizontal,
which ensures minimum disturbance to the bubble–floc agglomerate [18]. However, Longhurst and
Graham reported that, in theory, the baffle angle can range from 45° to 90° to the horizontal [15].
“Purac” have used a vertical baffle in the production of the “Flofilter” tank to avoid an eddying current during clarification and hydraulic congestion during filter backwash. But for the conventional
filter, they preferred to use an inclined baffle.
The depth of water below the water surface is found to vary across treatment plants, and greater
depths than those recommended by the Water Research Centre of 0.3–0.4 m are normally used [15].
In South Africa, the depth varies from 1.5 to 3.5 m, and in the United Kingdom it varies from 1.0 to
3.2 m [85]. This means there is still no agreement about the extent to which depth affects the optimization of design criteria. Regarding the width of the tank, it was observed that widths between
2.4 and 9.4 m are found in practice. However, Gregory and Zabel reported that tank widths are less
significant to hydraulic flow and are sometimes restricted by the sludge-removing device [11]. A
study carried out by Heinanen on the use of DAF for potable water treatment in Finland showed that
the design parameters for the process are still far from ideal and this has resulted in high construction costs [14]. He indicated that the situation could be avoided if research institutes had played an
important part in the design work.
Discussions with Noone [82] indicated that there is a need to investigate the optimum shape of
the tank. This means further investigations would be useful to justify the arrangements of the nozzles, the distance between the inlet and the baffle, the baffle angle, and the depth of water surface
from the baffle. Longhurst and Graham indicated that if the length of the tank runs only up to point
A (Figure 5.12), the tank may be too short, and the floc will not achieve its optimum flotation that
occurs at point B [15,87]. At point C, the tank is too long, which will cause the floc to settle down.
Noone indicated that even in Severn Trent Water, there is a range of tank sizes with varying
rectangular shapes and different arrangements of air nozzles, baffles, depths, and operational
Reaction
zone
Inlet zone
Air nozzle
Area
(separation zone)
A
B
C
FIGURE 5.12 Arrangements in flotation tank. (From Adlan, M.N., A study of dissolved air flotation tank
design variables and separation zone performance. PhD thesis, University of Newcastle Upon Tyne, 1998;
Murshed, M.F., Removal of turbidity, suspended solid and aluminum using DAF pilot plant. MSc thesis,
Universiti Sains Malaysia, 2007.)
170
Waste Treatment in the Service and Utility Industries
procedures, with no evidence to prove their effectiveness [82]. Thus, it is worth investigating these
parameters so that a better fundamental understanding can be developed regarding the optimization
of tank dimensions and the flotation process in practice. This could result in the saving of power,
chemicals, and operation times and in the development of standard design procedures.
Based on the comments by Noone, Adlan had carried out investigations at several water
treatment plants belonging to Severn Trent Water [88]. Acoustic Doppler Velocimeter with threedimensional down-looking probe was used in the investigation as it is ideal for measuring velocity
close to the bottom of the boundary layer. Turbidity data were monitored at the same points where
velocity data were taken in the separation zone of DAF tanks. Comparisons were made, and results
showed that the design of the rectangular DAF tank could be improved by reducing the length of
the separation zone without reduction in turbidity.
Downloaded by [Hamidi Aziz] at 16:04 26 September 2017
5.5 ADVANTAGES OF DAF APPLICATION IN WASTEWATER TREATMENT
The flotation technique has distinctive advantages over the conventional gravity settling technique for
the removal of low-density particles that have a tendency to float. Flotation techniques are classified
based on the methods of producing bubbles. The advantages of DAF are as follows [23,89–91,106–108]:
• Efficient at removing particles and turbidity, resulting in more economical filter designs. It
allows short detention time of about 5–10 min in flocculation tanks.
• Higher hydraulic loading rates can be used than in most settling processes.
• More efficient than sedimentation in removing low-density floc formed from coagulation
of Total Organic Compound (TOC).
• It allows lower coagulant dosages resulting in smaller chemical storage and lesser sludge.
• Smaller footprints with stacked flotation over filtration arrangement.
• Improved algae removal and cold water performance.
• Less sensitive to flow variations.
• Process flexibility through air loading.
5.6 APPLICATION OF DAF PROCESS IN WASTEWATER TREATMENT
DAF process is a physical process which has been used widely in wastewater treatment. DAF is
generally used as a combination process with coagulation. The industries that implement DAF for
their wastewater treatment process are paper mill, chemical-mechanical polishing (CMP) wastewater, meat industry, personnel care product, and seafood industry [5,92–95]. Oily wastewater or oil
refineries wastewater also use DAF for their wastewater treatment [90,97–101].
Zoubolis and Avranas applied coagulation and DAF process in treating simulated oily wastewater (using octane) with an initial concentration of 500 mg/L [96]. The coagulants used in this
study were anionic polyacrylamide, cationic (K-1384), and FeCl3. Sodium oleate (NaOl) was
used as a collector. It is a common anionic flotation collector used to enhance the floatability of
coagulated solids and emulsion droplets (flotation collector). The DAF system was operated as
a batch study using DAF jar-test with a pressure of 4–5 atm. and recycle ratio of 30%. Results
indicated that the use of polyacrylamide and cationic (K-1384) did not offer a good solution.
However, using FeCl3 to demulsify and increase the droplets size improved the droplet–bubble
adhesion and the overall DAF performance. The best removal rate for oil removal was obtained
at 100 mg/L Fe3+ with 50 mg/L NaOl at pH 6 and recycle ratio of 30%. The pH was optimum at
6 because at this pH, with the addition of Fe3+, the zeta potential of the hydrocarbon particle is
nearly zero, and the electrostatic barrier is greatly reduced, which thus improves the droplet–
particle attachment.
Al-Shamrani et al. carried out another study using DAF and coagulation in treating oily wastewater [25]. The study was carried out in a batch process. The sample was a solvent-refined petroleum
Downloaded by [Hamidi Aziz] at 16:04 26 September 2017
Dissolved Air Flotation (DAF) for Wastewater Treatment
171
distillate manufactured from crude petroleum oil. Alum was used as a coagulant agent at 100 mg/L
with pH 8. The DAF system was equipped with a packed saturator column with 90% efficiency, and
the recycle ratio was set at 10%. Two situations were observed in this study. First, at low concentration of alum with high working pressure for the saturator (80 psi) and high recycle ratio (20%),
only 72% of oil was removed. However, increasing the dosage up to 100 mg/L at pH 8 with the
same working pressure (80 psi) but at lower recycle ratio (8%), almost all the oil was removed. This
finding shows that the increase in alum concentration destroyed the protective action created by
the emulsifying agent (compound in oily wastewater). Simultaneously, the repulsive effects of the
electrical double layers (zeta potential) were reduced as the concentration of alum increased. This
results in the formation of larger droplets or particles through coalescence.
In 2003, Zouboulis et al. studied the removal of humic acid from synthetic landfill leachates as
a possible post-treatment stage after biological treatment using coagulation and DAF [100]. This
experiment used a column flotation. Samples were prepared with humic acid concentration from 50
to 300 mg/L, with 30% humic acid accounted for in the COD value. It means that after the biological treatment of the landfill leachate, the COD value ranges from 500 to 1000 mg/L, in which humic
acids account for around 30%. This compound is a very reluctant organic acid. The coagulation
process carried out prior to DAF used ethanol as frother and N-acetyl-N,N,N-trimethylammoniumbromide (CTAB) as a collector. The frother function is used to control the efficiency of bubble generation and to combine effectively with the collector. In the DAF system, the bubbles are produced
by injecting air from the bottom of the column through a microporous frit at a flow rate of 200 cm3/min.
According to Zouboulis et al., there are three main steps in the flotation mechanism:
1. Surface modification of the specific “species” (concerned compounds or contaminants) in
the water or wastewater using chemicals.
2. Contact and adherence of hydrophobic species with bubbles.
3. Separation of particles.
In this study, 70 mg/L of collector was used with 5 min flotation time resulting in 95% humic acid
removal. As a result, this study proved that the flotation process can be used as a post-treatment in
the landfill leachate. Besides, they discovered the parameters that control this process: (1) type and
dosage of frother, (2) dosage of collector, (3) pH solution, (4) ionic strength (concentration of other
salt), and (5) flotation time. Studies conducted by other researchers dealt with all these parameters,
except the ionic strength. In the study conducted by Zouboulis et al., two different salts were studied, the Na2SO4 and NaCl [100]. Increasing the concentration of these salts above 0.01 M decreases
humic acids removal. This proves that the ions (SO42− and Cl−) compete or interact with the cationic collectors. As a result, ions reduce the chances of humic acids to float. However, comparing
SO42− and Cl−, the Cl− effects are not significant and they decrease the overall percentage removal
of humic acids to 10% only. This may be due to the charges carried by SO42−, which is double, thus
reacting more with the collector compared to Cl− with one charge.
Another important point in this work is the zeta potential measurement at the optimum dosage.
Results indicate that reduction in zeta potential (from negative value to zero) increases in percentage
the removal of humic acids. However, beyond the optimum points, the results of zeta potential show
the opposite charge (positive). This proves that to obtain a good agglomeration, the zeta potential
of the concerned compound with collector or coagulant agent should be low or near zero.
Another study by Hami et al. showed the ability of coagulation and DAF in treating wastewater
in refineries [98]. The sample was collected from the effluent of the American Petroleum Institute
(API) separator. The coagulation process was carried out as a pre-treatment using a mixed coagulant containing alum, polyelectrolyte, and powered activated carbon (PAC). The suitable pH was in
the range 6.5–8.0. The DAF process was carried out in a laboratory as a pilot scale. Three different
pressures (2, 3, and 5 atm.) were chosen for this study, and the flow rate depended on the flow characteristics of pumps. For the DAF and recycle unit, the flow rate was 1.0–5.0 L/h and 1.0–2.0 L/h,
Downloaded by [Hamidi Aziz] at 16:04 26 September 2017
172
Waste Treatment in the Service and Utility Industries
respectively. In the first phase of the study, the DAF process as a single treatment did not offer
good COD and BOD removals. Only 19%–64% and 27%–70% of COD and BOD removals were,
respectively, obtained using DAF alone. However, increase in percentage removal was achieved
by a combination of coagulation and DAF. The COD and BOD removals were 72%–92.5% and
76%–94%, based on the initial concentration of 198 and 95 mg/L, respectively. Alum was believed
to reduce the zeta potential of the particle in the influent, which thus provoked the agglomeration
process, while the polyelectrolyte worked as a flocculant agent, and the PAC provided a site for the
particles to absorb, and subsequently the floc was removed by flotation. Another factor that was
addressed in this study was the flow rate through the nozzle in the flotation tank. Hami et al. found
that the increase in flow rate causes a decrease in percentage removal because when high flow rate
was applied, the resident was reduced (the bubbles do not have enough time to float all the particles).
Owing to this, the amount of effluent with lower contaminant concentration was reduced, which
resulted in low recycle ratio. Subsequently, it reduced the dispersed air bubbles due to lower gas
holdup.
Tsai et al. utilized nano-bubble flotation to treat CMP wastewater [92]. The experimental works
were carried out using laboratory and pilot scale experiments. The coagulants, also known as activators, used in this study were alum, ferric chloride, and PACL. The flocculant (collector) types
utilized after the coagulation process were CTAB and NaOl. In the DAF system, a special equipment named NBG (nano-bubble generator) was used to produce the nano-sized bubble. The recycled water was injected into the flotation cell through NBG at pressure around 7.7 atm. The bubble
size was around 30–5000 nm. Based on an experiment using the laboratory scale DAF unit, PACL
performs better than alum. However, a combination of FeCl3 with NaOl is superior to other combinations. Conversely, a combination of PACL–NaOl was used in the NBFT (nano-bubble flotation
technology) as a pilot scale due to some advantages of PACL in terms of cost and chemical characteristics. Experimental results using the NBFT process show that chemical cost was reduced four
times that of the conventional coagulation–sedimentation process. The addition of the collector
shows that it is able to absorb at the air–liquid interface and enhance the resistance of the bubble to
burst. The collectors provide an electrical potential to the bubble–particle and enhance the flotation
process. Another advantage of adding a collector (NaOl in this study) is that it is able to reduce the
bubble size, thereby increasing the flotation process. Moreover, the hydraulic retention time (HT),
bubble size, bubble quantity, and saturator pressure considerably influenced the NBFT process.
Based on this study, the optimum condition for NBFT process in treating the CMP wastewater was
50–60 mg/L of PACL and 5–10 mg/L of NaOl with 1 h HT, and 10%–20% recycle ratio. The pH
was not adjusted, and the raw pH value was 9.4. More than 95% of turbidity, total solids, and total
silica were removed.
de Nardi et al. focused on treating slaughterhouse wastewater using DAF process as laboratory
and pilot plant scale experiments [93]. Coagulation process was carried out prior to DAF using
PACL and anionic polyacrylamide. In the pilot plant scale experiment, the coagulation process
used 24 mg Al3+/L PACL with 1.5 mg/L anionic polymer, followed by DAF process using 100%
influent pressurization at 300 kPa. The lab-scale experimental process was conducted with and
without coagulation. The concentration of the coagulant and flocculant was the same as that in the
pilot scale experimental process. However, the pressure was higher compared with the pilot scale,
which was 450 kPa. Another difference between pilot and laboratory scale processes in this study
was the recycle ratio. In the laboratory scale experimental work, 40% recycle ratio was applied.
Results indicated that using the pilot-scale full pressurization, the removal for Suspended solids
(SS), oil and grease (O&G) was unsatisfactory at 28%–58% and 41%–57%, respectively. However,
in a lab-scale setup with recycle ratio, the removal is higher than the pilot plant scale experiment.
This could be due to the recycle ratio and pressure. As mentioned earlier, these two factors significantly affect the DAF efficiency in terms of the bubble size and bubble concentration. Comparison
of DAF with and without coagulation in the lab-scale experimental setup showed that the combination of coagulation and DAF gave higher removal than DAF as a single process. Around 54%–60%
Downloaded by [Hamidi Aziz] at 16:04 26 September 2017
Dissolved Air Flotation (DAF) for Wastewater Treatment
173
and 46%–74% removals of SS and O&G, respectively, were obtained when using DAF as a single
treatment, while a combination treatment (coagulation/DAF) gave 74% and 99% removal in their
respective order. This proves that chemical treatment is very important for improving DAF performance. Recycle ratio is an important factor for boosting the removal efficiency. Recycle ratio is also
efficient in controlling and varying the amount of air supplied.
de Sena et al. utilized coagulation/DAF as pre-treatment to treat the wastewater from the meat
industry [94]. Ferric chloride with 80 mg/L Fe3+ was used during the coagulation process, and the
DAF process was operated at 4.0 bars, with 5 min saturation time and 20% recycle ratio. Application
of coagulation/DAF in this study gave 60.5% and 78% removal of BOD5 and COD from the initial
concentration of 1500 and 2900 mg/L, respectively. This research work proves that this chemical or
physical process can be used as a pre-treatment process.
Another application of DAF was carried out in treating personnel care product wastewater collected from Unilever Masheq Company, Egypt [95]. Comparison was made between coagulation or
precipitation and coagulation/DAF. Three types of coagulants were tested. The coagulants concentration was 600, 800, and 700 mg/L, while the optimum pH was 8.23, 9.1, and 6.9 for ferric chloride,
ferric sulphate, and alum, respectively. The DAF process was operated at 4.0 bars. The coagulation/
precipitation process gave COD removal of around 75.8%, 77.5%, and 76.7% using ferric chloride,
ferric sulphate, and alum, respectively, and the BOD5 removals were 78%, 78.7%, and 74%, respectively. The coagulation or DAF process gives different outputs, such as differences in BOD removal.
Using the three types of coagulants is not significant. However, the COD removals were 71.5%,
67.7%, and 77.5% for ferric chloride, ferric sulphate, and alum, respectively. This experimental
study shows that the use of coagulation or DAF offers lower investment and running cost compared
with coagulation or precipitation, around 27.3% and 23.7%, respectively. The application of DAF is
summarized in Table 5.6.
5.7 APPLICATION OF DAF PROCESS IN LANDFILL LEACHATE TREATMENT
To date, the DAF is used more in industrial wastewater treatment for paper mill, meat processing,
and oil-based industries like POME, soap and oil refinery industry. The performance of DAF in
numerous research works has been summarized in Table 5.6. In 2000, Zouboulis et al. studied
humic acid removal from simulated leachate using DAF process [96]. However, based on the literature review, no study has been carried out for actual landfill leachate treatment using DAF process.
This indicates that there is a gap in the knowledge of DAF capability in leachate treatment. Based on
the advantages offered by DAF, such as higher hydraulic loading, allowing lower coagulant dosages,
being less sensitive to flow variations, and many more (described in Section 5.5), DAF process can
be a good alternative for landfill leachate treatment. Recently, study on semi-aerobic landfill leachate treatment using DAF was carried out by Palaniandy et al. and Adlan et al. [101,102].
The study carried out by Palaniandy et al. suggests a way to apply this method as a large-scale
application in landfill areas [101]. This research work investigates the application of DAF in leachate treatment with and without alum coagulation. Based on the study, a coagulation process must be
introduced to facilitate the destabilization of colloidal particles or emulsions. This is because DAF
process in leachate treatment without coagulation shows that the variations in percent removal of
turbidity, color, and COD were considerably low, which implies that the main pollutant in the leachate was in soluble organic and inorganic matter such as humic acid, fulvic acid, iron, sodium, potassium, sulfate, and chloride, [102,103]. However, with the addition of coagulant (alum), the percentage
removal of the studied parameters increased to 70%,79%, and 42% for color, COD, and turbidity,
respectively. Based on this study, DAF process is able to treat landfill leachate. Further research has
been carried out by Adlan et al. in this field by optimizing coagulation and DAF process in semiaerobic landfill leachate using response surface methodology (RSM). In this research work, ferric
chloride (FeCl3) was chosen to induce coagulation [102]. The results show 50%, 75%, 93%, and
41% reduction in turbidity, COD, color, and NH3–N, respectively. Comparing this research work
174
Waste Treatment in the Service and Utility Industries
TABLE 5.6
Summary of Previous Studies on DAF Process
Parameter Concern
Downloaded by [Hamidi Aziz] at 16:04 26 September 2017
Operating Parameter
Findings
Authors
Oil removal
•
•
•
•
•
100 mg/L Fe3+
50 mg/L NaOl
pH 6
Saturator pressure 4–5 atm.
Recycle ratio 30%
Zouboulis and
Avranas [96]
Oil removal
•
•
•
•
•
100 mg/L alum
pH 8
Saturator pressure (80 psi)
Recycle ratio (20%)
Packed saturator column
with 90% efficiency
Humic acid removal
• Ethanol as frother
• 70 mg/L CTAB as a collector
• Bubbles were produced by
injecting the air from the
bottom of the column
through a microporous frit,
at a flow rate of 200 cm3/
min.
• 5 min flotation time
COD and BOD from
refineries
wastewater
• Mixed coagulant contained
alum, polyelectrolyte, and
powered activated carbon
(PAC).
• Suitable pH is in the range
of 6.5 and 8.0.
• The flow rate depended on
the flow characteristics of
pumps; for DAF and recycle
unit, the flow rate was
1.0–5.0 L/h and 1.0–2.0 L/h,
respectively.
Results indicate that the use of
polyelectrolyte did not offer favorable
results. However, using the FeCl3 to
demulsify and increase the size of
droplets and improve the droplet–bubble
adhesion, the overall DAF performance
increased.
Increase in alum concentration destroyed
the protection action created by the
emulsifying agent (compound in oily
wastewater). Simultaneously, the
repulsive effects of the electrical double
layers (zeta potential) were reduced as
the concentration of alum increased.
Removal of humic acid was 95%.
This study proves that the flotation
process can be used as a post-treatment
in landfill leachate.
Increasing the concentration of salts
(SO42– and Cl–) above 0.01 M resulted in
the decrease in humic acids removal.
Comparison between SO42− and Cl−, the
Cl− effects was not significant, and it
decreased the overall percentage
removal of humic acids, which was only
10%.
To obtain a good agglomeration, the zeta
potential of the concerned compound
with collector or coagulant agent should
be low or near zero.
19%–64% and 27%–70% removal of
COD and BOD, respectively, using DAF
alone.
Combination of coagulation and DAF,
COD, and BOD removals were
72%–92.5% and 76%–94%,
respectively, from the initial
concentration of 198 and 95 mg/L,
respectively.
The alum was believed to reduce the zeta
potential of the particle in the influent;
thus, it provokes an agglomeration
process while the polyelectrolyte works
as a flocculant agent; the PAC provided
a place for the particles to adsorb, and
subsequently the floc was removed by
flotation.
Al-Shamrani
et al. [25]
Zouboulis
et al. [100]
Hami et al.
[98]
(Continued)
175
Dissolved Air Flotation (DAF) for Wastewater Treatment
Downloaded by [Hamidi Aziz] at 16:04 26 September 2017
TABLE 5.6 (Continued)
Summary of Previous Studies on DAF Process
Parameter Concern
Operating Parameter
Findings
Authors
Turbidity, total
solids, and total
silica removal from
CMP wastewater
• 50–60 mg/L of PACL
• 5–10 mg/L of NaOl
1 h HT
• 10%–20% recycle ratio
• pH was not adjusted, and the
raw pH value was 9.4
• NBG (nano bubble generator)
was used to produce the
nano-sized bubble.
• Pressure around 7.7 atm.
• Bubble size was around
30–5000 nm
This study indicated 40% greater removal
than the conventional coagulation–
sedimentation process.
More than 95% turbidity, total solids, and
total silica removal efficiencies were
obtained.
Tsai et al. [92]
O&G
SS from
slaughterhouse
wastewater
Pilot scale;
• 24 mg Al3+/L PACL
1.5 mg/L anionic polymer
• 100% influent
pressurization at 300 kPa
• Lab scale (without
coagulation);
• Pressure at 450 kPa
40% recycle ratio
• Lab scale (with
coagulation);
• 24 mg Al3+/L PACL
• 1.5 mg/L anionic polymer
• Pressure at 450 kPa
• 40% recycle ratio
Using the pilot-scale full pressurization,
the removal for SS and O&G was
unsatisfactory at 28%–58% and
41%–57%, respectively.
Using the lab-scale DAF process (without
coagulation), the removal was
54%–60% and 46%–74% for SS and
O&G, respectively.
Using the lab-scale DAF process (with
coagulation), the removal was 74% and
99% for SS and O&G, respectively.
de Nardi et al.
[93]
BOD5 and COD
from meat industry
wastewater
• Ferric chloride with
80 mg/L Fe3+
• Pressure at 4.0 bars
• 5 min saturation time
• 20% recycle ratio
60.5% and 78% removal for BOD5 and
COD from initial concentration of 1500
and 2900 mg/L, respectively.
de Sena et al.
[94]
BOD5 and COD
from personnel care
product wastewater
Coagulation/precipitation;
• 600, 800, and 700 mg/L,
and optimum pH was
8.23%, 9.1%, and 6.9%
for ferric chloride, ferric
sulphate, and alum,
respectively
Coagulation/DAF;
• 600, 800, and 700 mg/L,
and optimum pH was
8.23%, 9.1%, and 6.9%
for ferric chloride, ferric
sulphate, and alum,
respectively
• Pressure at 4.0 bars
COD removal;
75.8%, 77.5%, and 76.7% for ferric
chloride, ferric sulphate, and alum,
respectively.
BOD5 removal;
78%, 78.7%, and 74% for ferric chloride,
ferric sulphate, and alum, respectively.
COD removal;
71.5%, 67.7%, and 77.5% for ferric
chloride, ferric sulphate, and alum,
respectively.
BOD removal was not significant between
these three types of coagulants.
The use of coagulation/DAF offers lower
investment and running cost compared
with coagulation/precipitation, around
27.3% and 23.7%, respectively.
El-Gohary
et al. [95]
176
Waste Treatment in the Service and Utility Industries
with previous studies [102,104], coagulation/sedimentation offers a good removal in landfill leachate treatment. However, in terms of chemical usage, the coagulation/sedimentation process needs
double that required by the coagulation/DAF process. The proper determination of type and dosage
of chemicals will not only improve the process but also influence the running cost as proved by
El-Gohary et al. [95]. Based on the study carried out by El-Gohary et al. [95], the initial investment
and the running cost for coagulation and sedimentation are higher by 27.3% and 23.7%, respectively,
compared with coagulation/DAF. Besides, the land area required for coagulation/DAF is less by
30% compared with coagulation/sedimentation. As a result, coagulation/DAF is more economical
compared with coagulation/sedimentation in treating wastewater [95].
Downloaded by [Hamidi Aziz] at 16:04 26 September 2017
LIST OF NOMENCLATURE
A
D
μ
U
A
Σ
Р
рg
CD
Re
ν
R
C0
Ci
Vr
Q
Ah
V
Δρ
S
PB
xB
KB
CS
Р
K
F
ap
P
Cr
Ca
Rr
K
Area (m2)
Drag force (kN)
Dynamic viscosity (kg/m/s)
Terminal velocity (m/s)
Radius of sphere (m)
Density of sphere (kg/m3)
Density of liquid (kg/m3)
Density of gas (kg/m3)
Drag coefficient
Reynolds number
Kinematic viscosity (m2/s)
Ratio of total removal of solid concentration after flotation to the inflow solid
concentration = 1 − C0 Ci
Effluent suspended solids
Influent suspended solids
Rising velocity of a single particle/air bubble (m/s)
Flow applied to the flotation unit (m3/s)
Horizontal area of the unit (m2)
Volume of bubble (m3)
Difference of density between liquid and gas (kg/m3)
Area of bubble (m2)
Vapour pressure (Pa)
Mol. fraction of the solute
Constant
Concentration of air in saturated liquid (mg/L)
Absolute pressure (Pa)
Henry’s law constant (M/atm)
Efficiency factor (kPa/mg/L)
Concentration of air precipitated on reducing the pressure to atmospheric
(mg/L)
Saturator pressure in atmospheres
Concentration of air released in the tank (mg/L)
Concentration of air that remains in solution at atmospheric pressure (mg/L)
Recycle ratio
Influent saturation factor
Downloaded by [Hamidi Aziz] at 16:04 26 September 2017
Dissolved Air Flotation (DAF) for Wastewater Treatment
ϕb
psat
pw
PA
P0
R
T
Mw
KH
Vb
h
σ
da
G
Q
ƞ1
ƞ2
rp
Rb
pp
Gj
Vj
uj
t
V
μf
x, y
u
v
K
G
ƞ
u0
c0
d
ƞD, ƞI, ƞG
Pe
Rb
Ub
Df
dp
d
vo
kb
177
Bubble volume concentration
Saturated density of air
Density of water (gm/cm3)
Dissolved pressure (dyne/cm2)
Atmospheric pressure (dyne/cm2)
Gas constant (erg/K·mol)
Absolute temperature (K)
Molecular weight of water (g/g·mol)
Henry’s law constant (dyne/cm2)
Volume of air occupied by a single spherical bubble (cm3)
Depth from the liquid surface to the bubble (cm)
Surface tension (dyne/cm2)
Volumetric mean diameter of bubble (cm)
Volumetric flow rate of air generated under decrease in pressure (cm3/s)
Volumetric flow rate (cm3/s)
Collision efficiency
Attachment efficiency
Particle radius (cm)
Bubble radius (cm)
Particle density (gm/mL)
Body force acting on the particle
Particle velocity (cm/s)
Velocity the fluid would have at the position of the particle if no particle were there (cm/s)
Time (s)
Component of particle velocity (cm/s)
Fluid viscosity (kg/m/s)
Cartesian position coordinates (cm)
Component of velocity field due to bubble (cm/s)
Dimensionless component of particle velocity (cm/s)
Particle inertia parameter
Dimensionless settling velocity of particle (cm/s)
Single bubble collision efficiency
Water velocity (cm/s)
Suspended particle concentration upstream from the collector where the flow is
undisturbed by the presence of the grain
Grain diameter (cm)
Theoretical values for single collector efficiency
Peclet number
Bubble radius (cm)
Bubble rising velocity (cm/s)
Particle diffusivity (cm2/s)
Diameter of suspended particle (cm)
Diameter of collector or bubble (cm)
Approach velocity of fluid (cm/s)
Boltzmann’s constant for bubble
178
Downloaded by [Hamidi Aziz] at 16:04 26 September 2017
αpb
Nb
Ab
Np
D
VT
T
As
VH
Ac
W
L
Waste Treatment in the Service and Utility Industries
Attachment efficiency
Total number concentration of bubbles
Projected area of bubble (m2)
Particle number concentration
Effective depth of flotation chamber (m)
Vertical rise rate of suspended solids (m/s)
Detention time (s)
Surface area of flotation chamber (m2)
Horizontal velocity (m/s)
Cross-sectional area of flotation chamber (m2)
Width of the flotation chamber (m)
Effective length of the flotation chamber (m)
REFERENCES
1. Rodrigues, R.T. and Rubio, J. (2007). DAF-dissolved air flotation: Potential applications in the mining
and mineral processing industry. International Journal of Mineral Processing. Vol. 82, pp. 1–13.
2. Merrill, C.W. and Pennington, J.W. (1962). The magnitude and significance of flotation in the mineral
industries of the United States. In Froth Flotation 50th Anniversary Volume (Edited by Fuerstenau, D.W.),
Chapter 4. The American Institute of Mining, Metallurgical, and Petroleum Engineers, Inc., New York.
3. Travers, S.M. and Lovett, D.A. (1985). Pressure flotation of abattoir wastewaters using carbon dioxide.
Water Research. Vol. 19, pp. 1479–1482.
4. Krofta, M. and Wang, L.K. (1982). Potable water treatment by dissolved air flotation. Journal American
Water Works Association. Vol. 74, pp. 305–310.
5. Krofta, M., Wang, L.K., and Pollman, C.D. (1988). Treatment of seafood processing wastewater by dissolved air flotation, carbon adsorption and free chlorination. Proceedings of the 43rd Purdue Industrial
Waste Conference, Purdue University, West Lafayette, IN, pp. 535–550.
6. Wang, L.K. and Wang, M.H.S. (1989). Bubble dynamics and air dispersion mechanisms of air flotation process systems. Part A: Materials balances. Proceedings of the 44th Purdue Industrial Waste
Conference, Purdue University, West Lafayette, IN, pp. 493–504.
7. Van Puffelen, J., Buijs, P.J., Nunn, P.A.N.M., and Hijnen, W.A.M. (1995). Dissolved air flotation in
potable water treatment: The Dutch experience. Water Science Technology. Vol. 31, pp. 149–157.
8. Childs, A.R., Burnfield, I., and Rees, A.J. (1977). Operational experience with the 2300 m3/day pilot
plant of the Essex Water Company. Papers and Proceedings of the WRC Conference: Flotation for
Water and Waste Treatment, Medmenham.
9. Nickols, D., Moerschell, G.C., and Broder, M.V. (1995). The first DAF water treatment plant in the
United States. Water Science Technology. Vol. 31, pp. 239–246.
10. Edzwald, J.K. and Walsh, J.P. (1992). Dissolved air flotation: Laboratory and pilot plant investigation.
AWWA Research Foundation and AWWA, Denver, CO.
11. Gregory, R. and Zabel, T.F. (1990). Sedimentation and flotation. In Water Quality and Treatment: A
Handbook of Community Water Supply (Edited by Pontius, F.W.), 4th edition. McGraw Hill, New York,
pp. 367–453.
12. Kitchener, J.A. (1984). The froth flotation process: Past and present and future-in brief. In The Scientific
Basis of Flotation (Edited by Ives, K.J.). NATO ASI Series. Martinus Nijhoff Publishers, The Hague,
the Netherlands, pp. 3–52.
13. Lundgren, H. (1976). Theory and practice of dissolved air flotation. Journal of Filtration and Separation.
Vol. 13, pp. 24–28.
14. Heinanen, J. (1988). Use of dissolved air flotation in potable water treatment in Finland. Aqua Fennica.
Vol. 18, pp. 113–123.
15. Longhurst, S.J. and Graham, J.D. (1987). Dissolved air flotation for water treatment: A survey of operational units in Great Britain. Public Health Engineering. Vol. 14, pp. 71–76.
16. Zabel, T.F. (1978). Flotation. Twelfth International Water Supply Congress, Kyoto, Japan.
Downloaded by [Hamidi Aziz] at 16:04 26 September 2017
Dissolved Air Flotation (DAF) for Wastewater Treatment
179
17. Packham, R.F. and Richards, W.N. (1975). Water clarification by flotation 3: Treatment of Thames water
in a pilot-scale flotation plant. WRC Technical Report TR2, Medmenham, UK.
18. Zabel, T.F. (1985). The advantage of dissolved-air flotation for water treatment. Journal American
Water Works Association. Vol. 77, pp. 42–46.
19. Gregory, R., Zabel, T.F., and Edzwald, J.K. (1999). Sedimentation and flotation. In Water Quality and
Treatment: A Handbook of Community Water Supplies (Edited by Letterman, R.D.). McGraw-Hill,
New York, pp. 7.1–7.87.
20. Ho, C.C. and Chan, C.Y. (1986). The application of lead dioxide-coated titanium anode in the electroflotation of palm oil mill effluent. Water Research. Vol. 20, pp. 1523–1529.
21. Zabel, T.F. and Melbourne, J.D. (1980). Flotation. Development in Water Treatment-1. Applied Science
Publishers, London.
22. Barnes, D., Bliss, P.J., Gould, B.W., and Vallenrine, H.R. (1981). Water and Wastewater Engineering
Systems, Chapter 8. Pitman Books Ltd., London.
23. Zabel, T.F. and Hyde, R.A. (1977). Factors influencing dissolved air flotation as applied to water
clarification. Papers and Proceedings of WRC Conference: Flotation for Water and Waste Treatment,
Marlow.
24. Zabel, T. (1984). Flotation in water treatment. In The Scientific Basis of Flotation (Edited by Ives, K.J.),
NATO ASI Series. Martinus Nijhoff Publishers, The Hague, the Netherlands.
25. Al-Shamrani, A.A., James, A., and Xiao, H. (2002). Destabilisation of oil–water emulsions and separation by dissolved air flotation. Water Research. Vol. 36, pp. 1503–1512.
26. Edzwald, J.K. (2010). Dissolved air flotation and me. Water Research. Vol. 44, pp. 2077–2106.
27. Haarhoff, J. and Steinbach, S. (1996). A model for the prediction of the air composition in pressure
saturators. Water Research. Vol. 30, pp. 3074–3082.
28. Vesilind, P.A. (2003). Wastewater Treatment Plant Design. Water Environment Federation, New York.
29. HDR Engineering, Inc. (2001). Handbook of Public Water Systems. John Wiley & Sons, New York.
31. Harper, J.F. (1972). The motion of bubbles and drops through liquids. In Advances in Applied Mechanics
(Edited by Chia, S.Y.). Academic Press, New York, Vol. 12, pp. 59–129.
32. Li, W.H. and Lam, S.H. (1964). Principles of Fluid Mechanics. Addison-Wesley Publ. Co., Inc.,
Reading, MA.
33. Gaudin, A.M. (1939). Principles of Mineral Dressing, Chapter VIII. McGraw-Hill, London.
34. Batchelor, G.K. (1988). An Introduction to Fluid Dynamics. University Press, Cambridge, MA.
35. Shaw, D.J. (1991). Introduction to Colloid and Surface Chemistry, Chapter 1, 4th edition. ButterworthHeinemann Ltd, Oxford.
36. Clift, R., Grace, J.R., and Weber, M.E. (1978). Bubbles, Drops, and Particles. Academic Press, New York.
37. Chorlton, F. (1967). Textbook of Fluid Dynamics, Chapter 8. D. Van Nostrand Co. Ltd., London.
38. Packham, R.F. and Richards, W.N. (1972). Water Clarification by Flotation 1: A Survey of the Literature.
WRA Technical Paper TP87, Medmenham, UK.
39. Moore, W.D. (1965). The Velocity Rise of Distorted Gas Bubbles in a Liquid of Small Viscosity. Journal
of Fluid Mechanics 23(04):749–766.
40. Jameson, G.J. (1984). Experimental techniques in flotation. In The Scientific Basis of Flotation (Edited
by Ives, K.J.), NATO ASI Series. Martinus Nijhoff Publisher, The Hague, the Netherlands, pp. 53–78.
41. Shannon, W.T. and Buisson, D.H. (1980). Dissolved air flotation in hot water. Water Research. Vol. 14,
pp. 759–765.
42. Fair, G.M., Geyer, J.C., and Okun, D.A. (1968). Water and Wastewater Treatment Engineering, Vol. 2,
Chapter 26. John Wiley & Sons.
43. Vrablik, E.R. (1959). Fundamental principles of dissolved-air flotation of industrial wastes. Proceedings
of the 14th Industrial Waste Conference, Purdue University, West Lafayette, IN, pp. 743–779.
44. Krofta, M. and Wang, L.K. (1989). Bubble dynamics and air dispersion mechanisms of air flotation
process systems, Part B: Air dispersion. Proceeding of the 44th Purdue Industrial Waste Conference,
Purdue University, West Lafayette, IN, pp. 505–511.
45. Fukushi, K., Tambo, N., and Matsui, Y. (1995). A kinetic model for dissolved air flotation in water and
wastewater treatment. Water Science Technology. Vol. 31, pp. 37–47.
46. Edzwald, J.K., Malley, J.P, and Yu, C. (1990). A conceptual model for dissolved air flotation in water
treatment. Water Supply. Vol. 9, pp. 141–150.
47. Fukushi, K., Tambo, N., and Kiyotsuka, M. (1985). An experimental evaluation of kinetic process of
dissolved air flotation. Journal Japan Water Works Association. Vol. 607, pp. 32–41.
48. Levich, V.G. (1962). Physicochemical Hydrodynamics. Prentice-Hall, Inc., Englewood Cliffs, NJ.
Downloaded by [Hamidi Aziz] at 16:04 26 September 2017
180
Waste Treatment in the Service and Utility Industries
49. Howe, R.H.L. (1958). A mathematical interpretation of flotation for solid-liquid separation. In Biological
Treatment of Sewage and Industrial Wastes, Vol. 2 (Edited by McCabe, B.S. and Eckenfelder, W.W.).
Reinhold Publishing Corp., New York, pp. 241–250.
50. Karamanev, G.K. (1994). Rise of bubbles in quiescent liquids. Journal American Institute Chemical
Engineer. Vol. 40, pp. 1418–1421.
51. Gerrard, W. (1980). Gas Solubilities Widespread Applications, Chapter 1. Pergamon Press, Oxford, UK.
52. Lovett, D.A. and Travers, S.M. (1986). Dissolved air flotation for abattoir wastewater. Water Research.
Vol. 20, pp. 421–426.
53. Backhurst, J.R., Harker, J.H., and Porter, J.E. (1974). Problems in Mass Transfer. Edward Arnold Ltd,
London.
54. Atkins, P.W. (1994). Physical Chemistry, Chapter 7. Oxford University Press, Oxford, UK.
55. Takahashi, T., Miyahara, T., and Mochizuki, H. (1979). Fundamental study of bubble formation in dissolved air pressure flotation. Journal Chemical Engineering Japan. Vol. 12, pp. 275–280.
56. Ward, A.S. (1992). Dissolved air flotation for water and wastewater treatment. Process Safety and
Environmental Protection. Vol. 70, pp. 214–218.
57. Klassen, V.I. and Mokrousov, V.A. (1963). An Introduction to the Theory of Flotation. Butterworths,
London.
58. Edwards, W.M. (1984). Mass transfer and gas absorption. In Perry’s Chemical Engineers’ Handbook
(Edited by Green, D.W.), 6th Edition. McGraw-Hill, New York.
59. Eckenfelder, W.W. Jr., Rooney, T.F., Burger, T.B., and Gruspier, J.T. (1958). Studies on dissolved air
flotation on biological sludges. In Biological Treatment of Sewage and Industrial Wastes, Vol. 2 (Edited
by McCabe, B.S. and Eckenfelder, W.W.). Reinhold Publishing Corp., New York, pp. 241–250.
60. Bratby, J. and Marais, G.V.R. (1974). Dissolved air flotation. Filtration & Separation. Vol. 11, pp. 614–624.
61. Rykaart, E.M. and Haarhoff, J. (1995). Behaviour of air injection nozzles in dissolved air flotation.
Water Science Technology. Vol. 31, pp. 25–35.
62. Gulas, V., Benefield, R.L., Lindsey, R., and Randall, C. (1980). Design considerations for dissolved air
flotation. Water & Wastewater Sewage Works. Vol. 127, pp. 30–31, 42.
63. Ramirez, E.R. (1980). Comparative physicochemical study of industrial wastewater treatment by electrolytic, dispersed and dissolved air flotation technologies. Proceedings of the 34th Industrial Waste
Conference, Purdue University, West Lafayette, IN, pp. 699–709.
64. Jones, A.D. and Hall, A.C. (1981). Removal of metal ions from aqueous solutions by dissolved air flotation. Filtration & Separation, September/October, pp. 386–390.
65. Bratby, J. and Marais, G.V.R. (1975). Saturator performance in dissolved-air (pressure) flotation. Water
Research. Vol. 9, pp. 929–936.
66. Cassell, E.A., Kaufman, K.M., and Matijevic, E. (1975). The effects of bubble size on microflotation.
Water Research. Vol. 9, pp. 1017–1024.
67. Collin, G.L. and Jameson, G.J. (1976). Experiment on the flotation of fine particles: The influence of
particle size and charge. Chemical Engineering Science. Vol. 31, pp. 985–99.
68. Rovel, J.M. (1977). Experiences with dissolved air flotation for industrial effluent treatment. Papers and
Proceeding of the WRC Conference: Flotation for Water and Waste Treatment, Bucks.
69. Reay, D. and Ratcliff, G.A. (1973). Removal of fine particles from water by dispersed air flotation. The
Canadian Journal of Chemical Engineering. Vol. 51, pp. 178–185.
70. Gochin, R.J. (1990). Flotation, In Solid-Liquid Separation (Edited by Svarovsky, L.), 3rd Edition.
Butterworths, London, pp. 591–613.
71. Reay, D. and Ratcliff, G.A. (1975). Experimental testing of the hydrodynamic collision model of fine
particle flotation. The Canadian Journal of Chemical Engineering. Vol. 53, pp. 481–486.
72. Flint, L.R. and Howarth, W.J. (1971). The collision efficiency of small particles with spherical air bubbles. Chemical Engineering Science. Vol. 26, pp. 1155–1168.
73. King, R.P. (1982). Flotation of fine particles. In Principles of Flotation (Edited by King, R.P.), South
African Institute of Mining and Metallurgy, Johannesburg, South Africa, pp. 215–225.
74. Sutherland, K.L. (1948). Kinetics of the flotation process. Journal Physical Colloid Chemistry. Vol. 52,
pp. 394–425.
75. Woodburn, E.T., King, R.P., and Colborn, R.P. (1971). The effect of particle size distribution on
the performance of a phosphate flotation process. Metallurgical and Material Transactions. Vol. 2,
pp. 3163–3174.
76. Yao, K.M., Habibian, M.T., and O’Melia, C.R. (1971). Water and waste water filtration: Concept and
applications. Environmental Science and Technology. Vol. 5, pp. 1105–1112.
Downloaded by [Hamidi Aziz] at 16:04 26 September 2017
Dissolved Air Flotation (DAF) for Wastewater Treatment
181
77. Malley, J.P. Jr. and Edzwald, J.K. (1991). Concepts for dissolved air flotation treatment of drinking
waters. Aqua. Vol. 40, pp. 7–17.
78. Friedlander, S.K. (1967). Particle diffusion in low speed flows. Journal of Colloid Interface Science.
Vol. 23, pp. 157–164.
79. O’Melia, C.R. (1985). Particles, pretreatment and performance in water filtration. Journal of
Environmental Engineering. Vol. 111, pp. 874–890.
80. Katz, W.J. and Wullschleger, R. (1957). Studies of some variables which affect chemical flocculation when used with dissolved air flotation. Proceedings of 12th Purdue Industrial Waste Conference,
Purdue University, West Lafayette, IN, pp. 466–479.
81. Eckenfelder, W.W. Jr. and O’Connor, D.J. (1961). Biological Waste Treatment, Chapter 5. Pergamon
Press, London.
82. Noone, G. (1995). Personal Communication.
83. Bratby, J. and Marais, G.V.R. (1975). Dissolved air (pressure) flotation: An evaluation of inter-relationships
between process variables and their optimisation for design. Water S.A. Vol. 1, pp. 57–69.
84. Edzwald, J.K. (1995). Principles and applications of dissolved air flotation. Water Science and
Technology. Vol. 31, pp. 1–23.
85. Hyde, R.A. (1975). Water clarification by flotation-4, design and experimental studies on a dissolved air
flotation plant treating 8.2 m3/h of River Thames Water. WRC Technical Report TR13, Medmenham, UK.
86. Haarhoff, J. and Vuuren, L.R.J. (1995). Design parameters for dissolved air flotation in South Africa.
Water Science and Technology. Vol. 31, pp. 203–212.
87. Murshed, M.F. (2007). Removal of turbidity, suspended solid and aluminum using DAF pilot plant. MSc
thesis, Universiti Sains Malaysia.
88. Adlan, M.N. (1998). A study of dissolved air flotation tank design variables and separation zone performance. PhD thesis, University of Newcastle Upon Tyne.
89. Ponasse, M., Dupre, V., Aurelle, Y., and Secq, A. (1998). Bubble formation by water release in nozzle II.
Influence of various parameters on bubble size. Water Research. Vol. 32, pp. 2498–2506.
90. Al-Shamrani, A.A., James, A., and Xioh, H. (2002). Separation of oil from water by dissolved air flotation. Colloids and Surfaces A: Physicochemical and Engineering Aspects. Vol. 209, pp. 15–26.
91. Edzwald, J.K. (2006). Dissolved air flotation in drinking water treatment. Chapter 6 in Interface Science
in Drinking Water Treatment, Vol 10, eds. G. Newcombe and D. Dixon. Elsevier, New York.
92. Tsai, J.C., Kumar, M., Chen, S.Y., and Lin, J.G. (2007). Nano-bubble flotation technology with coagulation process for the cost-effective treatment of chemical mechanical polishing wastewater. Separation
and Purification Technology. Vol. 58, pp. 61–67.
93. de Nardi, I.R., Fuzi, T.P., and Del Nery, V. (2008). Performance evaluation and operating strategies of
dissolved-air flotation system treating poultry slaughterhouse wastewater. Resources, Conservation and
Recycling. Vol. 52, pp. 533–544.
94. de Sena, R.F., Moreira, R.F.P.M., and Jose, H.J. (2008). Comparison of coagulants and coagulation aids
for treatment of meat processing wastewater by column flotation. Bioresource Technology. Vol. 99,
pp. 8221–8225.
95. El-Gohary, F., Tawfik, A., and Mahmoud, U. (2010). Comparative study between chemical coagulation/
precipitation (C/P) versus coagulation/dissolved air flotation (C/DAF) for pre-treatment of personal care
products (PCPs) wastewater. Desalination. Vol. 252, pp. 106–112.
96. Zouboulis, A.I. and Avranas, A. (2000). Treatment of oil-in-water emulsions by coagulation and
dissolved-air flotation. Colloids and Surfaces A: Physicochemical and Engineering Aspects. Vol. 172,
pp. 153–161.
97. Bensadok, K., Belkacem, M., and Nezzal, G. (2007). Treatment of cutting oil/water emulsion by coupling coagulation and dissolved air flotation. Desalination. Vol. 206, pp. 440–448.
98. Hami, M.L., Al-Hash, M.A., and Al-Doori, M.M. (2007). Effect of activated carbon on BOD and
COD removal in a dissolved air flotation unit treating refinery wastewater. Desalination. Vol. 216,
pp. 116–122.
99. Poha, P.E., Onga, W.Y.J., Laub, E.V., and Chonga, M.N. (2014). Investigation on micro-bubble flotation
and coagulation for the treatment of anaerobically treated palm oil mill effluent (POME). Journal of
Environmental Chemical Engineering. Vol. 2, pp. 1174–1181.
100. Zouboulis, A.I., Jun, W., and Katsoyiannis, I.A. (2003). Removal of humic acids by flotation. Colloids
and Surfaces A: Physicochemical and Engineering Aspects. Vol. 231, pp. 181–193.
101. Palaniandy, P., Adlan, M.N., Aziz, H.A., and Murshed, M.F. (2010). Application of dissolved air flotation (DAF) in semi-aerobic leachate treatment. Chemical Engineering Journal. Vol. 157, pp. 316–322.
182
Waste Treatment in the Service and Utility Industries
Downloaded by [Hamidi Aziz] at 16:04 26 September 2017
102. Adlan, M.N., Palaniandy, P., and Aziz, H.A. (2011). Optimization of coagulation and dissolved air flotation (DAF) treatment of semi-aerobic landfill leachate using response surface methodology (RSM).
Desalination. Vol. 277, pp. 74–82.
103. Aziz, H.A., Alias, S., and Adlan, M. (2007). Colour removal from landfill leachate by coagulation and
flocculation processes. Bioresource Technology. Vol. 98, pp. 218–220.
104. Lo, I.M.C. (1996). Characteristics and treatment of leachates from domestic landfills. Environment
International. Vol. 22, pp. 433–442.
105. Ghafari, S., Aziz, H.A., and Bashir, M.J.K. (2010). The use of poly-aluminum chloride and alum for the
treatment of partially stabilized leachate: A comparative study. Desalination. Vol. 257, pp. 110–116.
106. Bondelind, M., Sasic, S., Bergdahl, L. (2013). A model to estimate the size of aggregates formed in a
dissolved air flotation. Applied Mathematical Modelling. Vol. 37, pp. 3036–3047.
107. Lakghomi, B., Lawryshyn, Y., and Hofmann, R. (2015). A model of particle removal in a dissolved air
flotation tank: Importance of stratified flow and bubble size. Water Research. Vol. 68, pp. 262–272.
108. Zhanga, Q., Liu, S., Yang, C., Chen, F., Lu, S. (2014). Bioreactor consisting of pressurized aeration
and dissolved air flotation for domestic wastewater treatment. Separation and Purification Technology.
Vol. 138, pp. 186–190.
View publication stats
Büşra Sarı