Dissolved air flotation (DAF) to improve the reuse of

water in mining operations

Yendery Ramrez
Department of Chemical and Mineral Process Engineering, Universidad de Antofagasta, Chile
School of Industrial Engineering and Management, Lappeenranta University of Technology, Finland

Gustavo Lara
Department of Physics, Universidad de Antofagasta, Chile

Luis Moreno
Department of Chemical Engineering and Technology, Royal Institute of Technology, Sweden

Antonio Garca
Department of Metallurgical Engineering, Universidad Catlica del Norte, Chile

Luis Cisternas
Department of Chemical and Mineral Process Engineering, Universidad de Antofagasta, Chile

ABSTRACT
The northern Chile is a zone with large needs of water sources, at present, enforces are dedicated to
the use of seawater in the mining operation, which is expensive since the mining operations are
located at significant height and large distance from the coastline. Therefore, an economical
alternative could be to reduce the water consumption by means of reuse, which requires the
removal of impurities (particles or chemicals) of the water before reuse. This may be accomplished
by Dissolved Air Flotation (DAF), which is a clarification process used for removing particles in
water treatment. The use of DAF for water treatment and reuse (or recycling) is rapidly increasing
in the mining sector.
A DAF reactor for particle removal was modelled by using computational fluid dynamics (CFD). I n
order to validate the model, a pilot DAF reactor is under construction, in which experiments will be
carried out and the results compared with the CFD simulations. The validated model could be then
used in optimization and operation of similar reactors working at other scales, conditions, or
configurations; e.g., a multi- step DAF, for saving energy. Basically, the process consists of a
coagulation-flocculation pre-processing followed by a cleaning process using DAF. Water
containing air dissolved is introduced at high-pressure and air micro-bubbles are formed when the
overpressure is released. Micro-bubbles collide with the particles forming bubble-particle bounds,
which travel to the surface of the cell forming a particle-laden foam that is removed from the
system. The CFD model uses an Eulerian description of the phases: the continuous water phase and
the two discrete phases, namely the bubbles without particles and the bubbles bounded to particles.
The system is solved for the velocity-fields, the pressure, and the air volume fraction; from these
parameters the particle removal efficiency is then obtained.

Keywords: Dissolved Air Flotation (DAF); Computational Fluid Dynamics (CFD); Water Treatment

1
INTRODUCTION
Dissolved air flotation (DAF) is used in water and wastewater treatment and basically consists of a
coagulation-flocculation pre-treatment followed by a cleaning process, in which water containing
air dissolved is introduced at high-pressure and air microbubbles are formed when the
overpressure is released, the bubbles then collide with the particles forming bubble-particle bounds,
which travel to the surface of the cell forming a particle-laden foam that is removed from the
system. The process is applied to the removal of particulate material; e.g., low-density flocs
(Bondelin, et al., 2010). DAF is an alternative to sedimentation in the removal of suspended solid
particles, metallic ions, microorganism and macromolecules (Kurama, et al., 2010).
The northern Chile is a zone with large needs of water sources, at present, enforces are dedicated to
the use of seawater in the mining operation, which is expensive since the mining operations are
located at significant height and large distance from the coastline. In the desalination of seawater,
DAF is used in some situations when high level of algae are found in the seawater; e.g., during the
red tides events (red algae bloom) (Petry, et al., 2007; Peleka and Matis, 2008). The minerals industry
is being driven to save freshwater and minimize mine water discharge (Liu, et al., 2013). Therefore,
an economical alternative could be to reduce the water consumption by means of reuse, which
requires the removal of impurities (particles or chemicals) of the water before reuse. The use of DAF
for water treatment and reuse (or recycling) is rapidly increasing in the mining sector, aqueous
solutions produced during hydrometallurgical operations require suitable treatment in order to
increase low concentrations of dissolved constituents and to make economically feasible their
subsequent recovery, hence constituting a situation similar to the wastewater treatment
(Zouboulis & Matis, 1997).
As shown in Table 1, DAF is applied in several processes in the mining industry for water cleaning
and posterior reuse. Zouboulis and Matis (1997) studied in laboratory-scale experiments the
removal of soluble ionic species from diluted aqueous solutions using a combination of sorption
and flotation. In the first step the ionic species are sorbed by fine or ultrafine particle-size sorbents,
which are removed by DAF. From the foam containing the ionic species, the ionic species may be
recovered. Therefore, if natural sorbents are used the load to the environmental were negligible. In a
similar process, adsorbing colloid flotation, metal ions are removed by adsorption on a
precipitate, which acts as a carrier. This process is used in Chile for effective separation of
Molybdenum from CuMo concentrate filtrates. In the process, suspended solids (SS) are firstly
separated by a rougher stage and Mo ions adsorbed on iron hydroxide precipitate are removed in a
cleaner stage, which occurs at pH about 5 (Rubio et al. 2002). Other application of DAF is ore
flotation assisted with microbubbles for the recovery of fine mineral particles (<13 m) where
microbubble addition shows improvements in particle recovery (2%) and in flotation kinetics
(Rodrigues & Rubio, 2007).
Dissolved air flotation may also be applied for cleaning of aqueous effluents generated in mining
activities; e.g. Acid mine drainage (AMD). Cadorin et al. (2007) experimentally studied the removal
of sulfate and Fe and Mn ions from AMD in coal and metal sulfides mines. They are precipitated at
pH 12 using lime and aluminum salts; the sulfate precipitates as ettringite and the metals as
hydroxides. They showed that DAF may effectively remove the solids formed.
Al-Zoubi and Al-Thyabat (2012) show the combined use of DAF and nanofiltration (NF) in cleaning
of wastewaters from a mine in Jordan, where about 10 000 m 3 of wastewater are discarded daily
containing about 1 000 ton solids, of which 200 tons are of P 2O5. They use column flotation and
DAF, however, the recovery of column flotation was low, only 20%. A better recovery was obtained

2
when using DAF (83%). In a posterior paper (Al-Zoubi & Al-Thyabat, 2012), the solid free
wastewater was purified using DAF and NF. SS is almost complete removed from the wastewater,
while the removal of chloride and sulfate is about 50%. The effluent treated by DAF was then filtrated
by NF. The removal of chloride was slightly increased, while that sulfate was totally removed.
Table 1 Examples of DAF applications in the mining industries and the wastewater treatment

Compound Recover Recovery

System Type Water Type Reference
(Removed) Efficiency
Algae Continuous 27-53% Lake freshwater Chu, et al. (2011)
Municipal Pouet &
Aluminum Continuous 70-80%
wastewater Grasmick (1995)

Batch and Kurama, et al.

Ammonium ions 88-99,21% Surface water
Continuous (2010)

Zouboulis, et al.
Cadmium cations Batch >95% - (1997) Esmaeili,
Plating et al. (2014)
Chromium - 98 %
wastewater Zamboulis, et al.,
Copper, Zinc and 2004
Chromium (VI) Batch >95% - Yalcin, & Byers
Fine particles Cu-Ni Continuous Cu 4 X % (2006)
ore (<37 m) Ore Pulp Yalcin, et al.,
Air-Aragon Ni 3.1 -23 %
Fine particles of iron (2002) Chuang,
ore (<37 m) Continuous >90% SiO2 Ore Pulp et al. (2002)
Fluoride-containing Semiconductor Menezes, et al.,
wastewater Bench Scale >95% (2011)
wastewater
Heavy metals Fe(III) Acid mine
Batch 94 % Karagzel, (2010)
and Mn(II) drainage
Iron containing Albite, Fe-Min Oliveira, & Rubio
minerals (Fe-Min) Batch 76 % (2009) Tassel,
and slimes
Isopropyl xanthate Synthetic et al. (1997)
Batch >90%
ions solutions Oliveira, et al.
Laboratory (1999)
Mercury >98% Process water
unit Feris, et al. (2004)
Oil
Batch 88-98% Synthetic solution Bensadok, et al.
Oils, dyes and metal (2007)
Batch and Synthetic and
ions >75% Al-Shamrani, et
Continuous industrial water
al. (2002)
Oil-water emulsion Synthetic
Batch 71 % Al-Thyabat & Al-
emulsions
Zoubi, (2012)
Oil-water emulsion Synthetic
Bench Scale >99,3%
wastewater Stoica, et al. (1995)
Phosphate 86,4% with
Bench Scale Wastewater Phoochinda &
20,6% P2O5
Radioactive White (2003)
substance 226Ra(II) - >91,4% Wastewater
Scenedesmus
quadricauda (algae) Batch 80-90% Surface water

3
The most of the development in DAF has been carried out by experimental work, but several
authors have also used modelling; e.g., Behin and Bahrami (2012) used computational fluid dynamics
(CFD) for modelling of the flow pattern in the flotation tank. Its results were compared with
experiments through of the residence time distribution curves, finding a good agreement. Bondelind,
et al. (2010) using single- and two-phase models of DAF in 2D and 3D simulations finding that a 2D
model cannot resolve successfully the flow in the contact zone, but major flow characteristics may be
adequately described.
The aim of the paper is the study of DAF, applied to the separation of particulate material in order
to clean wastewater for posterior reuse. CFD will be used in the modelling of the process and the
results of the simulations will be compared with observations in a tank a lab-scale. The strategy is
to validate the model in lab scale, once that the model is adequately validated, this could be used
for modelling of flotation system in industrial scale. Therefore, the experimental setup is
sufficiently flexible for accommodate different design.

METHODOLOGY
As described above, the methodology is based in the development of a CFD model to take into
account the movement of the water and bubbles in the reactor and the interaction of the bubbles
with the impurities (particles) to be removed from the wastewater. Several approaches exist with
different complexity degrees; the most complex modelling is used when we are interested in
describe how the location of specific bubbles changes with time. In that case, the Navier-Stoke
equations are applied for the liquid and bubbles in detail. When the detailed location of given
bubbles is not important, as in our case, the model calculates the average volume fractions occupied
by the gas phases instead of determining the location of each phase in detail (Buscaglia et al., 2002).

The model
We use Multiphase Eulerian model with three phases: the water phase, the clean air-bubbles phase,
and the loaded air-bubbles phase. The particulate material will be considered as a scalar and this
does not interact with the water, but is the source of loaded air bubbles, and the sink of the clean
air-bubbles. The partial differential equations are solved for the velocity-fields, the pressure, and
the volume fraction using the CFD code FLUENT 15. The three phases are treated as interpenetrating
continua and are described by phasic volume fractions. The inlet boundary conditions are the inlet
velocities for water and air. The outlet boundary conditions are the pressure at the outlet zones. The
model considers two water in-flows; one for the water to be treated and another for the water
containing the dissolved air, from which the air will be released. This inlet is also used for the air
bubbles, which are assumed to be 0.05 mm in diameter. The implicit scheme is used to solve the model
equations.
In the Viscous Model, we consider turbulence: realizable k- model per phase, with near-wall scalable
wall functions and options such as differential viscosity model, swirl dominated flow, and curvature
correction are used. The system is formed by water with volume fractions w, clean air- bubbles with
volume fraction a, and loaded air-bubbles with volume fraction L. In addition to the bubbles exist
small particles (impurities) whose concentration at the inlet is C0 = 1 kg/m3, and decreases along
its trajectory through the seawater, as impurities are colliding with clean air bubbles. We defined
the scalar = CC0, which is the relative concentration at each point of the fluid.

4
Impurities particles are small and in low concentration, therefore, the hydrodynamic of the system
is not altered. The loaded air-bubbles consist of attaching a clean air-bubble with a particle.
Mass Conservation Equation
( ) ( ) (1)
( ) ( ) ( )( ) (2)
( ) ( ) ( )( ) ( ) (3)

Scalar Equation
( ) ( ) ( ) (4)

In that equation, the K constant is

(5)
P is the probability of adherence if the particle collides with the bubble, Acol cross-section area for
impact, vrel relative velocity between bubble and particle. The mechanisms that cause the collisions
are: flotation movement of the bubbles, gravitational movement of the particles, and turbulence. Each
of these mechanisms has its own value of P and vrel.

The equipment
The DAF cell was designed to allow changes in its dimensions; e.g., the width of the contact zone and
the type of perforated plate at the separation zone bottom may be varied. The height of the contact
zone may also be varied. Figure 1 shows a schematic picture of the DAF cell. The depth of the cell is
0.05 m; other dimensions are shown in the figure.
Wastewater

0.03 m Treated water

Sludge

1.08 m
Separation
1.03 m zone
0.86 m
Contact
zone 0.82 m

Perforated
plate

Water with
dissolved air 0.05 m

0.03 m 0.30 m

5
 Figure 1 A schematic picture of the DAF cell, cell depth is 0.05 m

The wastewater enters the flotation cell at the top and descends along the inlet zone. At the bottom
the wastewater enters the contact zone through a slot and is mixed with the air micro-bubbles. In
the contact zone, the mixture is brought upward. The mixture flows horizontally over the shaft wall
and enters the separation zone. Cleared water is removed at the bottom via a perforated plate. The
tank arrangement is capable of operating at surface loadings up to 5.7 m/h.

RESULTS AND DISCUSSION

Simulations were performed considering a flow of seawater of 1.2 L/min, with a concentration of
impurities of C0 = 1 kg/m3. The performance of the reactor is determined by measuring the impurity
concentration at the reactor outlet, which is expressed by the scalar . In a quasi-stationary regime
the gas in the compressor has a composition of nitrogen higher than that found in the air, since the
oxygen is more soluble in water than the nitrogen. For this reason, at the initial stage, the oxygen
dissolves in water to a higher ratio than its composition ratio. Therefore, the dissolution of nitrogen
in seawater is used as the coefficient of gas dissolution in seawater, 16.6 mg/(kg atm), for a
temperature of 20C. Impurities are particles 0.02 mm in diameter with a density of 2000 kg/m 3.
In these simulations, the pressure used in the saturator is 400 KPa and bubble injection rate is 1 10-7
kg/s, which corresponds to an injection ratio bubbles/impurities of 1.04. The energy consumed in
the injection is 8.25 KWh/m3 water.
Figure 2 shows the volume fraction of clean bubbles, bubbles loaded with impurities, and the
impurity scalar concentration in the flotation cell. Logarithmic scales are used in Figures 2a and 2b.
Figure 2c uses linear scale.

Figure 2 Volume fractions and concentration of impurities: (a) volume fraction of clean bubbles, (b) bubbles
loaded, and (c) impurity scalar concentration in the vertical plane in the middle of the flotation cell

6
To achieve this bubble injection rate, 7.3% of the treated water should be re-injected to the reactor.
This bubble injection rate cleans up 72.8% of impurities. At the clean water outlet, the flow rate of
clean bubbles is negligible, but the flow of loaded bubbles is 24.9% of the injected bubbles, because
its mass is 105 times that of a clean bubble. Results for other bubble injection rates, is showed in
Table 2.
Table 2 Some quantities in terms of bubble injection rates
Bubble injection Cleaning Reinjected Energy cost Loaded bubbles in
rate (kg/s) reached water (W h/m3) water outlet
1.0 10-7 72.8 % 7.3 % 8.25 24.9 %
1.5 10-7 94.9 % 11.0 % 12.4 26.5 %
2.0 10-7 97.2 % 14.6 % 16.5 31.8 %

Figure 3 shows the seawater velocity in magnitude (left) and as vector (right). It may be observed
that the water distribution is very uneven distributed. In general, it is thought that this water
distribution is not favorable to obtain a good cleaning of the water with impurity particles. It is
possible to improve this, reducing the inflow of water, but also modifying the geometry of the reactor
keeping a high water flowrate.

Figure 3 Velocity of seawater in the DAF reactor

As shown in Table 2, to improve the cleaning reached, the bubble injection and the recirculated water
have to be significatively increased. Therefore, a possible alternative for improving the
performance and to do the process energy efficient is to work in two stages. Higher percentages of
cleaning may then be obtained working in two stages than in a single stage.

CONCLUSION
The reuse of water in mining activities is an important issue to reduce the water demand, which is
critical in the northern Chile. Dissolved Air Flotation, DAF, is a convenient method to water

7
cleaning, and its use has benn strongly increased in the last decades as was shown in Table 1. There
are a lot of applications in which DAF is used for cleaning water in the mining sector.
A DAF reactor for particle removal was modelled by using computational fluid dynamics (CFD). The
model was applied to a case where very small particles are removed from the water using DAF.
It was found that in order to improve the cleaness degree reached i the process, the air bubble injection
and the recirculated water have to be significatively increased. Therefore, a possible alternative
for improving the performance and to do the process more energy efficient is to work in two stages.
Higher cleanness percentages may be then obtained working in two stages than working in
one single stage.
Once that the CFD model were validated using our experimental DAF cell at pilot scale will be used
to simulated a DAF in two stages. The pilot reactor is under construction since part of the
equipment was not opportunely delivered.

ACKNOWLEDGEMENTS
The authors thank CONICYT and the regional government of Antofagasta for supporting through
the PAI program, project Anillo ACT 1201.

NOMENCLATURE
Cross-section area for impact.
Concentration of mpurities at the inlet.
K Frequency of collisions between bubbles and particles.
P Probability of union between bubbles and particles.
t Time.
m Mass.
V Volume.
Relative velocity.
Velocity vector
Volume fraction
Relative concentration of impurities.
Density.

Sub index
a Clean bubbles.
b Bubbles.
L Load bubbles
p Particles.
w Seawater.

8
REFERENCES
Al-Shamrani, A. A., James, A., & Xiao, H. (2002) Destabilisation of oilwater emulsions and separation by
dissolved air flotation, Water Research, 36(6), 1503-1512.
Al-Thyabat, S., & Al-Zoubi, H. (2012) Purification of phosphate beneficiation wastewater: Separation of
phosphate from Eshydia Mine (Jordan) by column-DAF flotation process, International Journal of Mineral
Processing, 110, 18-24.

Al-Zoubi, H. S., & Al-Thyabat, S. S. (2012) Treatment of a Jordanian Phosphate Mine Wastewater by Hybrid
Dissolved Air Flotation and Nanofiltration, Mine Water and the Environment, 31(3), 214-224.
Behin J. and Bahrami S. (2012) Modelling an industrial dissolved air flotation tank used for separating oil from
wastewater, Chemical Engineering and Processing: Process Intensification, 59, 1-8.

Bensadok, K., Belkacem, M., & Nezzal, G. (2007) Treatment of cutting oil/water emulsion by coupling
coagulation and dissolved air flotation, Desalination, 206(1), 440-448.

Bondelind M., Sasic S., Kostoglou M., Bergdahl L., Pettersson T.J.R. (2010) Single- and two-phase numerical
models of dissolved air flotation: Comparison of 2D and 3D simulations, Colloids and Surfaces A:
Physicochemical and Engineering Aspects, 365, 137-144.
uscaglia, , ombardelli, , arc a, M H ( ) Numerical modeling of large-scale bubble plumes
accounting for mass transfer effects, International journal of multiphase flow, 28(11), 1763-1785.

Cadorin, L., Carissimi, E., & Rubio, J. (2007) Avances en el tratamiento de aguas cidas de minas, Scientia et
Technica, 1(36).
Chu, W. H., Gao, N. Y., Templeton, M. R., & Yin, D. Q. (2011) Comparison of inclined plate sedimentation and
dissolved air flotation for the minimisation of subsequent nitrogenous disinfection by-product
formation, Chemosphere, 83(5), 647-651.

Chuang, T. C., Huang, C. J., & Liu, J. C. (2002) Treatment of semiconductor wastewater by dissolved air
flotation, Journal of environmental engineering, 128(10), 974-980.

Esmaeili, A., Hejazi, E., & Hassani, A. H. (2014) Removal of Chromium by Coagulation-Dissolved Air
Flotation System Using Ferric Chloride and Poly Aluminum Chloride (PAC) as Coagulants, Water,
Air, & Soil Pollution, 225(10), 1-8.

Feris, L. A., De Leon, A. T., Santander, M., & Rubio, J. (2004) Advances in the adsorptive particulate flotation
process, International Journal of Mineral Processing, 74(1), 101-106.

Karagzel, C. (2010) Selective separation of fine albite from feldspathic slime containing colored minerals (Fe-
Min) by batch scale dissolved air flotation (D ), Minerals Engineering, 23(1), 17-24.

Kurama, H., Karagzel, C., Mergan, T., & elik, M. S. (2010) Ammonium removal from aqueous solutions by
dissolved air flotation in the presence of zeolite carrier, Desalination, 253(1), 147-152.

Liu W., Moran C.J., Vink S. (2013) A review of the effect of water quality on flotation, Minerals Engineering, 53,
91-100.
Menezes, C. T., Barros, E. C., Rufino, R. D., Luna, J. M., & Sarubbo, L. A. (2011) Replacing synthetic with
microbial surfactants as collectors in the treatment of aqueous effluent produced by acid mine drainage,
using the dissolved air flotation technique, Applied biochemistry and biotechnology, 163(4),
540-546.
Oliveira, R. C. G., Gonzalez, G., & Oliveira, J. F. (1999) Interfacial studies on dissolved gas flotation of oil droplets
for water purification, Colloids and Surfaces A: Physicochemical and Engineering Aspects, 154(1),
127-135.

9
Oliveira, C. R., & Rubio, J. (2009) Isopropylxanthate ions uptake by modified natural zeolite and removal by
dissolved air flotation, International Journal of Mineral Processing, 90(1), 21-26.
Peleka N.E, Matis K.A. (2008) Application of flotation as a pretreatment process during desalination,
Desalination, 222, 1-8.
Petry M., Sanz M.A., Langlais C., Bonnelye V., Durand JP., Guevara D., Nardes W.M., Saemi C.H. (2007) The
El Coloso (Chile) reverse osmosis plant, Desalination, 203, 141-152.
Phoochinda, W., & White, D. A. (2003) Removal of algae using froth flotation, Environmental technology, 24(1),
87-96.
Pouet, M. F., & Grasmick, A. (1995) Urban wastewater treatment by electrocoagulation and flotation, Water
science and technology, 31(3), 275-283.

Rodrigues, R. T., & Rubio, J. (2007) DAFdissolved air flotation: Potential applications in the mining and
mineral processing industry, International Journal of Mineral Processing, 82(1), 1-13.

Rubio, J., Souza, M. L., & Smith, R. W. (2002), Overview of flotation as a wastewater treatment technique,
Minerals Engineering, 15(3), 139-155.
Stoica, L., Catuneanu, R., & Filip, G. (1995) Decontamination of solutions containing radioactive substances by
dissolved air flotation, Water Research, 29(9), 2108-2112.
Tassel, F., Rubio, J., Misra, M., & Jena, B. C. (1997) Removal of mercury from gold cyanide solution by
dissolved air flotation, Minerals engineering, 10(8), 803-811.
Yalcin, T., Byers, A., & Ughadpaga, K. (2002) Dissolved gas method of generating bubbles for potential use in
ore flotation, Mineral Processing and Extractive Metallurgy Review, 23(3-4), 181-197.

Yalcin, T., & Byers, A. (2006) Dissolved gas flotation in mineral processing, Mineral Processing and Extractive
Metallurgy Review, 27(2), 87-97.
Zamboulis, D., Pataroudi, S. I., Zouboulis, A. I., & Matis, K. A. (2004) The application of sorptive flotation for
the removal of metal ions, Desalination, 162, 159-168.

Zouboulis, A. I., & Matis, K. A. (1997) Removal of metal ions from dilute solutions by sorptive flotation,
Critical reviews in environmental science and technology, 27(3), 195-235.
Zouboulis, A. I., Matis, K. A., Lanara, B. G., & Loos-Neskovic, C. (1997) Removal of cadmium from dilute
solutions by hydroxyapatite. II. Flotation studies, Separation Science and Technology, 32(10), 1755-1767.

10