University
stringclasses 19
values | Text
stringlengths 458
20.7k
|
|---|---|
Virginia Tech | reagents, particularly collectors, may substantially interfere with frother sorption to coal; and 3)
the experimental conditions (e.g., mixing, effective contact time) used here may not be
representative of plant conditions. Considering these, the sorption mechanisms of frothers to coal
and tailings particles is deserving of further study. If, for example, frothers are identified which
sorb strongly to coal through flotation and dewatering, this may have significant implications for
reducing fouling of process circuits in closed water systems, as well as reducing environmental
releases through tailings impoundments. For frothers that do not sorb to and remain with coal,
novel water treatment strategies may be devised to remove these reagents from water prior to
recycling or environmental discharges.
Figure 2.2: Surface tension versus varying dosage levels of frother and coal
5.2 Collector Adsorption
DRO results (i.e., the residual DRO in the clear water fractions of tested coal slurries) are
presented in Tables 2.3 and 2.4 for all test conditions. The most striking observation is that there
is some low level of DRO in every test, despite the addition of even large amounts of coal (i.e.,
32 |
Virginia Tech | 10% solids). For instance, tests 12 and 13 clearly show that at relatively high solids content (i.e.,
1 and 10%, respectively) and very low dosages of diesel (i.e., <1 mg/L; or 0.17 and 0.017 lb/ton,
respectively), about 0.3-0.4 mg/L DRO remains in the water fraction of the slurry. Moreover, the
level of DRO does not change dramatically between tests, considering the extreme changes in
diesel and coal dosages. In test 20, for example, which had the same amount of coal but nearly
300x more diesel added than test 12, the DRO concentration was only about 2x higher than that
of test 12 (i.e., 0.79 vs. 0.42 mg/L, respectively). And in test 22, which had the same amount of
coal but nearly 3000x more diesel added than in test 13, the DRO concentration was only
increased by about 6x (i.e., 1.92 vs. 0.31 mg/L, respectively). These results seem to indicate that
a small amount of diesel (~0.3 mg/L or less) is always soluble in the water, but that the coal
particles have a very high adsorption capacity for the diesel that is not dissolved. Another factor
that may have been at play here is the possible presence of colloidal matter in the water fraction
of the slurries; if diesel sticks to the colloids, it would likely be measured as DRO. However, it is
important to note that, no matter what the reason, these tests indicate that a small amount of
diesel will effectively partition with water in a flotation circuit.
Figure 2.3 highlights other specific observations in the collector partitioning tests. In the
far left plot, the effect of solid-liquid separation technique on the results is shown. The three tests
(#s 3-5) were conducted using identical slurries (i.e., % coal solids and diesel dosage), but one
was centrifuged, one was filtered, and the other was centrifuged and then filtered. DRO
concentrations in the clear water fraction from each of these tests were all within about 15% of
each other – a reasonable range for preliminary tests – and it was concluded that the solid-liquid
separation methods did not substantially impact partitioning results (e.g., by sorption of diesel to
the filter paper).
33 |
Virginia Tech | Figure 2.3: Diesel sorption test results
The middle plot of Figure 2.3 shows the results from six tests to determine the
reproducibility of the test and analytical methods used here. Tests 14-16 show DRO measured
three separate times (i.e., in triplicate) from a single sample. The results for these tests are within
about 20% of each other and suggest that the analytical method is fairly reproducible. Likewise,
tests 17-19 show DRO measured from samples from three separate, but identical tests. In this
case, the results are within about 18% of each other and indicate that the test method is also
reproducible.
In the right plot of Figure 2.4 are the results from three tests conducted to determine the
effect of proportionally similar coal and diesel additions (i.e., tests at 1, 5 and 10% solids, each
with a diesel dosage of 50 lb/ton coal). Since the diesel was dosed on the basis of coal weight, it
seems intuitive that DRO concentrations should have been similar between these tests; instead,
with increasing additions of coal, less diesel actually sorbed. One possible explanation for this
phenomenon may be that with more coal in the slurry, particles are sticking to each other or
being bridged together by diesel such that there are effectively fewer sorption sites available. For
tests where coal content remained constant (e.g., tests 7-9) but diesel dosage was varied,
measured DRO in the water did increase with a substantial increase in diesel dosage – although
not proportionally. For instance, in tests with 5% Hagy Seam coal (-100 mesh), DRO was
roughly equal for diesel dosages of 0.25 and 1.0 lb/ton (i.e., 0.50 and 0.53 mg/L), but essentially
doubled when the diesel dosage was raised to 10 lb/ton (i.e., to 0.95 mg/L).
It was further observed that the ash content of coal appears to affect diesel sorption. At
equal slurry contents and diesel dosages (i.e., 5% solids, and diesel dosages of 10 or 50 lb/ton),
the Pocahontas Seam coal (~16% ash) sorbed about 2-2.5x as much diesel as the Hagy Seam
34 |
Virginia Tech | coal (~35% ash) (see Table 2.4). This is likely because coal has a higher affinity for diesel than
ash does. It is difficult to assess whether or not the sized Hagy Seam coal (100 x 150 mesh)
behaved differently than that which was only ground (-100 mesh), since just one test condition
was repeated between the first and second set of tests (i.e., tests 9 and 24; 5% coal and diesel
dosage of 10 lb/ton); however, the DRO results for these tests were practically very similar.
In terms of real preparation plants, the results of the collector partitioning tests presented
here indicate that, as expected, most diesel should partition with the coal. However, some
(presumably soluble) diesel may well remain in the process water – eventually being sent to
tailings impoundments or being recycled back through the plant. While no Federal water quality
standards currently exist for DRO, some states have set levels of concern at 0.05 mg/L (e.g.,
through reporting levels for diesel spills or contamination from underground storage tanks) (DEP
2002). The topic of soluble DRO, including the relative solubility of specific diesel compounds
and potential remediation strategies, is deserving of additional research.
6. Conclusions
Processing reagents used in coal preparation have a wide range of potential
environmental fates, as well as implications for preparation circuits that are designed or revised
to utilize closed water systems. The preliminary test work presented in this paper confirms that
common frother and collector reagents are not likely to partition completely to a single fraction
of the process slurry. Instead, the partitioning phenomena are complex, and appear to depend on
many operating variables including coal and reagent characteristics and dosages.
To gain a better understanding of the ultimate fates of these reagents and related impacts,
further work should focus on determining the mechanisms by which various reagents may
associate with solid and liquid fractions of coal slurries. Moreover, work is needed to elucidate
strategies for controlling/optimizing reagent partitioning, or treatment of affected process
streams.
7. Acknowledgments
The authors would like to acknowledge the Appalachian Research Initiative for
Environmental Science (ARIES) and US Department of Energy (under grant no. DE-AC22-86-
TC91221) for funding experimental work described here. Views, opinions or recommendations
35 |
Virginia Tech | OPTIMIZATION OF AIR-INJECTION SPARGERS
FOR COLUMN FLOTATION APPLICATIONS
Viviana Ramirez Coterio
ABSTRACT
Column flotation cells have become the most popular machine design for
industrial applications that require high purity concentrates. The superior metallurgical
performance of column cells can be largely attributed to their unique geometry which
readily accommodates the use of froth washing systems. This unique feature allows
column cells to provide impressive levels of metallurgical performance closely
approaching the ultimate separation curve predicted using flotation release analysis.
Another very important feature of column cells is the gas sparging system.
Unfortunately, field studies suggest that gas injector systems are not always optimized.
Two possible reasons for this unfavorable status are (i) improper design of the sparging
system and (ii) poor operation practices employed by plant operators. In light of these
issues, an experimental study was performed to develop a better understanding of the
effects of various design and operating variables on the performance of a commercial
gas sparging system. The data collected from this work was used to develop
operational guidelines that plant operators can employ to improve column performance
and to correct flaws in the design of their gas sparging systems. |
Virginia Tech | OPTIMIZATION OF AIR-INJECTION SPARGERS
FOR COLUMN FLOTATION APPLICATIONS
Viviana Ramirez Coterio
GENERAL AUDIENCE ABSTRACT
Column flotation cells have become the most popular separation device designed
for industrial applications requiring the concentration of wanted or unwanted mineral
from the rest material in a pulp. To achieve separation, an air sparging device is
required to produce bubbles in the flotation cell. In column flotation operations sparging
devices generate small bubbles into the cell to carry the desired mineral to the surface
for later recovery and processing. However, field studies suggest that air injector
systems are not always optimized. Reasons contributing to the lack of optimization are:
(i) ineffective internal design of the sparging system, and (ii) poor operation techniques
by the industrial processing plants.
The objective of this present study is to better understand sparging performance
into the column cell and how to optimize sparging systems more effectively. To achieve
this end, data for gas-water injection rate, froth addition, and inlet-pressure are collected
and analyzed. Based on the data collected and its analysis, a guideline to better
operational practices that plant operators can employ to improve column performance
was developed. Furthermore, the correction of flaws in the design of the sparging
devices was possible translating in an improvement in bubble generation inside the
flotation cell. |
Virginia Tech | ACKNOWLEDGMENTS
Foremost, I would like to express my sincere gratitude to my advisor, Dr. Gerald
Luttrell, for his patience, encouragement, and continues guidance throughout the course
of this work. In spite of his busy life, he always took the time to help and guide me; not
only as a professor but also as a friend.
Additionally, I want to thank Dr. Mike Mankosa and his team, Eriez Flotation
Division, not only for funding this project but also for their support, direction, and
comprehension. Furthermore, I want to thank this group for leading my work on
interesting and diverse projects while being a summer intern in their company.
Also, I would like to express my gratitude to Bob Bratton and Jim Waddell for
their assistance, support, and orientation that allowed for this project to be completed.
Last, but not the least, I am grateful to my best friend Susan Dar for her
continuous encouragement, guidance, care, and belief in me. Above all, I
wholeheartedly thank my fiancé John Marulanda for his unconditional love, patience
and extreme support.
iv |
Virginia Tech | 1. INTRODUCTION
1.1 Preamble
Column flotation is a separation method that was born within the mining industry
for the recovery of fine particles, normally less than 100 microns in diameter. This
process is now employed in several industries, e.g., in the civil industry for soil recovery
and wastewater treatment, in the paper recycling industry for paper deinking, and in the
petrochemical industry for oil-water separation. In the mining industry, column flotation
is well suited because this technology offers better efficiency than others technologies in
the selective separation of particles. The high selectivity is attributed to the employment
of wash water that is added to the top of the column to reduce the hydraulic entrainment
of fine hydrophilic particles. Column flotation machines make use of a variety of gas
aeration systems for bubble generation. As stated by Rubinstein (1995), optimal
performance of the aerator system is imperative as this is responsible for bubble
generation. The bubble size generated by the aerator system is considered to be one of
the most important parameters affecting column flotation performance. However, few
aerator systems offer the possibility to monitor, control, and even less, predict the
bubble size generated within the column. This unfavorable situation can create a
number of issues within the column operation, including poorer concentrate grade and
lower recovery, among others.
A basic understanding of column flotation operation is essential in order to
recognize the critical function of the gas sparging system. A column flotation cell is
basically a cylindrical vessel with a large height-to diameter ratio. Gas is introduced
near the bottom of the cell through a gas distributor system. The dispersed bubbles rise
1 |
Virginia Tech | in countercurrent fashion to the downward flow of feed slurry. Through the introduction
of gas into the pulp, air bubbles selectively adhere to naturally or chemically altered
hydrophobic particles. The bubbles carry these hydrophobic particles to the surface
where they are recovered as a froth phase. The remaining hydrophilic particles stay in
the pulp phase and are removed as a tailings stream through a discharge valve located
at the very bottom of the column. As mentioned before, the use of wash water at the top
of the column improves selectivity by washing undesirable material that may get
hydraulically entrained into the froth phase.
A mayor constraint on column flotation capacity is froth overloading. The carrying
capacity of the froth depends on the bubble surface area available for bubble-particle
attachment. The bubble surface area, and hence carrying capacity, can be increased by
reducing the average size of bubbles for a given gas flow rate. Efficient and proper air
sparging performance is vital to the success of column flotation operation as an
increase in bubble size decreases gas holdup or gas volume in the flotation pulp, thus
decreasing the probability of bubble-particle collision and attachment. The introduction
of finer bubbles to the cell also improves flotation kinetics and increases the total bubble
surface area flux (Laskowski, 2001). Based on this concept, it can be said that spargers
become the most important device for column flotation operations; therefore, without
burping or surging, they should produce the maximum rate of bubble surface area
throughout the column (Kohmuench, Mankosa, Wyslouzil, & Luttrell, 2009). This
objective can be achieved by controlling the gas dispersion performance of the sparger,
which is heavily dependent on the proper balance of gas and water flows and the
design of the sparger.
2 |
Virginia Tech | Among the available aeration systems to be employed in column flotation
operations, the market offers both static and dynamic sparging systems. In the static
sparging system category, porous bubblers are widely used in several industries.
However, due to plugging problems, porous bubbles often cannot be employed in
mineral processing operations due to plugging issues. In the mining industry, porous
spargers are therefore usually confined to laboratory testing or pilot plant evaluations.
In the dynamic sparging system category, the market offers several options including
jetting, Microcel and CavTube spargers. These three types of sparging systems are
widely employed and accepted in the mining industry for column flotation operations
due to their high efficiency, low cost and reliability, to name a few advantages. Moreover
research has shown that dynamic spargers, which employ high energy dissipation to
disperse gas within the column, are the most suitable devices for the control and
prediction of bubble size in column operations.
1.2 Problem Statement
The performance of column flotation is strongly influenced by the effectiveness of
the gas sparging system. Unfortunately, field studies suggest that gas injector systems
used for column sparging are not always optimized. This unfavorable condition can
create a number of issues within the concentrator, including lower recoveries, poorer
metallurgical upgrading, decreased capacities, increased circulating loads, higher
reagent consumption and inefficient energy usage. In order to avoid these problems and
to obtain an optimal level of performance, the sparging system must be properly
designed, installed, operated, and maintained. An effective sparging system should
create small and uniform bubbles throughout the column (Yoon and Luttrell, 1989).
3 |
Virginia Tech | The present study is focused on the evaluation and optimization of the very
popular “SlamJet” gas sparger manufactured by the Eriez Flotation Division (EDF). Field
studies suggest that this type of gas injector system is often not fully optimized, which
can translate into poor column performance. Therefore, in order to have peak column
performance, the sparging system must to be properly designed and well operated. In
terms of design, this level of performance requires the upfront selection of the proper
number of sparger nozzles, the best choice of nozzle diameters, and the best sparger
distribution pattern across the column cross-sectional area. Because in the
manufacturing industry there is uncertainty as to how to design a sparging system that
can bring optimum results to the mineral processing operations, this study focused on
the development of guidelines that can be adapted by plant operators to improve
sparging system performance and, at the same time, can positively impact column
performance in terms of throughput capacity and separation efficiency. The techniques
and modifications proposed from this work can also be used to improve future designs
of gas injector sparging systems.
1.3 Objectives
The main objective of this project is to review the important criteria that govern
sparging system operation. The investigation also reviews how the design of these
sparging systems can influence column flotation performance. This study primarily
focuses on one type of gas dispersion system, the SlamJet® sparger, which has shown
increased popularity in the mining industry for mineral processing applications. The
SlamJet® sparger, which is manufactured by the Eriez Flotation Division, operates by
passing compressed gas (and often a small amount of water) through a small discharge
4 |
Virginia Tech | nozzle. Fluid turbulence created by the exiting gas (and water if used) disperses and
distributes small bubbles into the flotation pulp.
The main aims of this study are:
To collect data required to improve sparging systems designs.
To determine gas/water flow rates, inlet pressure, frother addition and
frother type for optimum sparging system operation.
To better understand the performance of jetting-type sparger systems in
column flotation.
1.4 Literature Review
1.4.1 Froth Flotation
Froth flotation is a physical method that relies on the naturally or chemically
altered hydrophobicity of certain minerals. This method is used to selectively separate
valuable minerals from unwanted gangue. In the mining industry, froth flotation is
typically used as the last stage of the mineral recovery/concentration system. It is used
to recover or upgrade materials that conventional gravity or magnetic separators cannot
recover due to the very fine particle size. Froth flotation allows the economic recovery of
valuable minerals from low-grade ores that were not possible to obtain decades ago.
Froth flotation cannot be possible without the introduction of air bubbles into the
flotation pulp. The froth flotation concept relies on the ability of air bubbles to adhere to
hydrophobic mineral surfaces. The bubble-particle aggregates rise to the surface of the
flotation pulp where they are later skimmed off as froth to make the separation. The
remaining unwanted material is then evacuated from the flotation machine as a tailing
stream. Froth flotation methods can also be employed to recover naturally hydrophilic
5 |
Virginia Tech | minerals. This is possible by altering the mineral particle surface from hydrophilic to
hydrophobic through chemical treatment so that air bubbles can attach to the mineral
and the separation can take place. The ability to change the mineral/material surface
through chemical treatment expanded the use of froth flotation in the mining industry
and also into new non-mining industries including water treatment, deinking of paper,
removal of organic contaminants in the dairy and beer industries, the remediation of
contaminated soils in the civil field, as well as other industries (Kantarcia, Borakb and
Ulgen, 2004). To increase hydrophobicity, or make a hydrophilic mineral hydrophobic,
fatty acids and oils were first employed as reagents during the early years of flotation
technology development. Now, a wide range of chemicals, including collectors, frothers,
activators, depressants, and pH regulators, are commonly used as a complement to
enable the flotation process and to increase the recoverability of valuable materials.
In 1869, William Haynes introduced the concept of flotation to separate sulfides
from gangue using oils. This process was called bulk oil flotation. The separation was
possible by bubbles generated through three different methods: (i) the entrainment of
air during mixing, (ii) the reduction of pressure to generate bubbles, and (iii) the addition
of sulfuric acid to create carbon dioxide bubbles (Fuerstenau, Jameson and Yoon,
2007). Later, in 1877, the Bessel brothers patented what is known today as froth
flotation to concentrate graphite minerals. They innovated the industry by using
nonpolar oils and by generating bubbles through the buoyancy of water to raise graphite
flakes to the surface. In 1896, Frank Elmore, in conjunction with his brother Stanley and
father William, developed, commercialized, and installed the first industrial-sized
flotation process to concentrate sulfide minerals in The Glasdir copper mine in North
6 |
Virginia Tech | Wales. The process patented by Frank in 1989 was not froth flotation, but it used oil to
agglomerate pulverized sulfides and bring them to the surface by buoyancy (Fuerstenau
et al., 2007). Although Frank’s technology successfully improved the separation of
sulfides from non-sulfides, new research realized the importance of bubbles in the froth
flotation process. Therefore, the flotation process was independently reinvented in other
places, especially in Australia at the beginning of 1900. Some of the new Australian
inventors were Charles Vincent and Guilleame Daniel Delprat.
Further improvements in the flotation process were accomplished throughout
history, but two flotation methods are very well established in the mining industry today
for mineral processing: the conventional mechanical cell and the column cell.
1.4.1.1 Conventional Mechanical Cell
The conventional mechanical flotation cell consists of a tank equipped with an
impeller and stator mechanism. The rotating impeller is located in the lower part of the
cell or tank. The air required to generate bubbles is introduced through a small-diameter
orifice near the impeller or through an orifice inside of the impeller (Rubinstein, 1995). In
the flotation cell, three distinct zones are observed during operation: (I) the turbulent
zone, (II) the quiescent zone, and (III) the froth zone. According to Miskovic (2011), “the
rotating action of the impeller in the turbulent zone (I) provides the energy necessary to
keep particles in suspension, enables the generation of small bubbles, and maintains
the hydrodynamic conditions needed for efficient bubble-particle interaction” (Miskovic,
2011). In Zone II, entrained gangue particles are separated or liberated from the
aggregates. In addition, Zone II helps to maintain the froth in a stable state. Zone (III) is
7 |
Virginia Tech | needed for optimum operation, and it does not offer the possibility for air flow control
(Sastri, 1998).
Figure 1.2 Possible configuration of Rougher – Scavenger – Cleaner Flotation Circuit
1.4.1.2 Column Cell
Column flotation depends on the principle of mass separation in a countercurrent
flow of air and slurry, which is ideally used in the flotation of fine (<100 microns)
particles (Kohmuench et al., 2009). Some of the characteristics that distinguish the
column flotation cell from the mechanical cell are its shape, which is cylindrical and
taller (up to 16 meters in height) (Kohmuench, Yan and Christodoulou, 2012), its bubble
generation system, and its use of wash water (Dobby & Finch, 1990).
Column flotation is the most recent major innovation in flotation equipment. Its
first design dates from 1919, when M. Town and S. Flynn developed a countercurrent
flow of slurry and air in a cylindrical tank. Inside the tank, previously conditioned pulp
was continuously fed into the middle part of the cylinder. Pressurized air, required for
bubble generation, was generated from a cloth aerator, or sparger, located at the
bottom of the cylinder (Rubinstein, 1995). As a result of problems such as particles’
sedimentation at the bottom of the apparatus and clogging of the air sparging system (a
9 |
Virginia Tech | cloth aerator or sparger), the use of the column cell was not popular. It was not until the
mid-sixties when researchers, P. Boutain and R. Tremblay, began to investigate the
mass separation process that occurred on a countercurrent slurry and air in a column.
The column flotation developed by these researchers, initially intended for the chemical
industry, is known as the “Canadian Column” and today is widely used in the mining
industry for mineral processing. The operational principle of the “Canadian Column” is
the same as that developed by Town and Flynn: the material/slurry, previously
conditioned with reagents, is fed into the column from the middle part of it. Once in the
column, the slurry encounters an ascendant stream of air bubbles rising from the bottom
of the column and generated by pressurized air from the sparger system.
The first commercial column cell installed at a mineral processing plant was used
to clean molybdenum ore in 1981 at Les Mines Gaspe in Quebec, Canada. After its
successful application to clean molybdenum, the column flotation apparatus became
accepted, and its use widely expanded in the late 1980’s through early 1990’s for the
roughing stage of sulfide and gold ores; the cleaning stage of copper, lead, zinc, and tin;
and for ash removal from coal (Rubinstein, 1995).
Throughout the investigation of the column flotation process, it has been found
that countercurrent flow provides a better condition for bubble/particle attachment,
which is governed by relative velocity, contact time, and inertia forces. The optimum
relative velocity for boarding, or attachment, to occur was found by F. Dedek. He
discovered that the optimum collision occurs under the following conditions: a relative
velocity of bubble and particles in the countercurrent of 10 – 12 cm/s, a bubble size of
1.5 – 2.5 mm, and a slurry superficial flow rate of 2 cm/s (Rubinstein, 1995). The joined
10 |
Virginia Tech | condition of the countercurrent flow of slurry and air reduces the bubble rise velocity.
These conditions increase retention time and reduce gas requirements, which in turn
improve the performance of the cell. With the absence of an impeller, which generates
high turbulence, the inertial forces that cause bubble/particle detachment are negligible.
In other words, as Rubenstein explains, “in a countercurrent, the probability of
bubble/particle collision is higher because of the large aerated volume of the cell and
the long distance the particle and bubble have to travel along the column height”
(Rubinstein, 1995).
The reduced cross-sectional surface area of a column cell benefits froth stability
and the formation of a deep froth bed. Having a deep froth bed facilitates the washing of
undesirable impurities from the floated product in the bubble swarm by the wash water,
which enters from the top of the cell. The primary advantage of having wash water at
the top of the cell is the superior separation performance it offers to the column cell
compared with the conventional mechanical cell (Kohmuench et al., 2012). Introducing
wash water from the top of the cell allows it to permeate through the froth zone,
removing dirty and nonselective entrainment of particles trapped between the bubbles.
Furthermore, it improves the stability and movement of the froth, allowing a relatively
deep (up to 1.5 meters) froth bed to be utilized. The deep froth promotes upgrading and
ensures good distribution of the wash water. Figure 1.3 schematically illustrates a
column cell with its main components and zones previously mentioned.
11 |
Virginia Tech | D is the particle diameter in the froth
p
D is the bubble diameter in the froth
b
Q is the gas flow rate
g
According to the equation above, carrying capacity can be increased if the
superficial bubble surface area rate is increased. Increased carrying capacity can be
attained by raising the aeration rate or by reducing bubble size. Studies have shown
that in the normal range of operation, air rate and column diameter have only a marginal
effect on carrying capacity (Sastri, 1996). Therefore, it can be inferred that optimal
carrying capacity can be achieved when the compressed gas system used in the
column is conducted at the maximum air velocity providing the minimum average
bubble size.
1.4.2 Sparging Systems
The aeration or sparging system, also known as the bubble generator device, is
the heart of the process in column cells, according to Rubinstein in his book, Column
Flotation. The service life, operational costs, and economical parameters of flotation
columns are tied to the design and operation of the device (Rubinstein, 1995). Proper
design and performance of the sparging system is essential for column flotation, as
spargers dictate and control bubble size, rise velocity, and air distribution. Hence,
spargers dictate both the radial and axial hold-up profiles as well as the liquid phase
flow patterns which translate to better flotation column performance (Kulkarni and Joshi,
2011). There are two methods of aeration systems. The first method is the internal air
system which is placed near the bottom of the column to directly inject air. The second
method is the external air system where gas and the liquid/slurry are introduced into the
13 |
Virginia Tech | column via external contacting. This type of sparger is used to aerate the moving slurry,
which is pumped from the bottom of the flotation cell and is recirculated as a pulp-air
mixture. External spargers present a great advantage in the column flotation process
compared to internal spargers since they can be maintained and repaired while the
column is in operation. In addition, external spargers are easily operated. These
advantages have expedited the development of column flotation.
In 1914, the first sparger devices were made to operate a pneumatic flottyoin
column from porous material such as filter cloth and perforated rubber (Rubinstein,
1995). At that time, researchers and operators used a perforated metal frame wrapped
in a woolen cloth. The air was then introduced to the slurry through the covered frame.
In the early stages of sparger development (early 1900’s) only internal spargers were
used, which all suffered from (i) plugging due to particles and/or precipitates, (ii)
improper gas distribution requiring large numbers of spargers to maintain bubble sizes
below 2-3mm, (iii) poor reliability due to tearing and deterioration with use, and (iv) the
need to shut down column operation to change them (Finch, 1994). All of these early
drawbacks forced operators and developers to improve sparging technologies;
however, the more significant improvements only in occurred the last few decades. As
sparging technologies improved, the popularity of column flotation significantly grew.
Thus, it is important to note that conditions present in the laboratory setting completely
differ from conditions at industrial settings. These differences restricted low-pressure
internal spargers for use only in the laboratory and for pilot test units (Kulkarni and
Joshi, 2011).
14 |
Virginia Tech | Although the overall goal of air sparging is the same for both internal and
external spargers; the design, sparging method, and features vary substantially
between the sparger types. In the mining industry, three types of spargers dominate the
mineral processing field. These are porous spargers, air injectors, and dynamic
spargers. A brief explanation on the design and operation of these three types is
provided bellow to illustrate their advancements and differences in the mining industry.
1.4.2.1 Static Sparging Systems. Porous Spargers
Perforated plates or pipes, sometimes covered by a porous filter cloth or
perforated rubber, were the first type of sparging system used in the early 1900’s to
operate pneumatic column flotations. With porous materials, bubble generation occurs
by the formation of individual bubbles at each orifice. The use of porous materials offers
finer bubble sizes if operated at low pressure. Furthermore, since bubbles created by
this type of gas distributor are numerous and relatively small, the gas-liquid interfacial
area is greater, offering more efficient mass transfer (Kazakis, Mouza, and Paras,
2008). Porous spargers also are less costly and can be reclaimed through washing
(Rubinstein, 1995). Perforated and/or porous spargers come in a variety of designs
including perforated pipes, frames, rings, grids, and plates/sieves.
While these types of spargers can be used in mineral processing operations, the
perforations (or holes) must be large enough to overcome clogging caused by the high
concentration of solids in the column’s bottom and the long operation times demanded
in industrial settings. Unfortunately, in practice, the ability to reduce the hole size to
minimize the average bubble size while eliminating fouling, is still an impossibility.
Therefore, this inability confines perforated spargers to be implemented for laboratory
15 |
Virginia Tech | testing and not in mineral processing applications at an industrial scale. Another
disadvantage of perforated spargers that makes them inadequate for mineral
processing operations at an industrial scale is the need to shut down flotation
operations for repair, and the potential for slurry to permeate into the air system
(Rubinstein, 1995).
To compensate for the lack of control in generating optimum bubble sizes,
researchers developed various forms of porous spargers; however, persistent fouling
also confined them to only laboratory scale work. Figure 1.4 shows a perforated plate
and perforated pipe/tube sparger configuration. Here, it can be seen that the perforated
plate sparger embraces the full cross sectional area of the column, while the perforated
tubes sparger configuration is designed to achieve the necessary air distribution
(Kulkarni and Joshi, 2011).
Figure 1.4 Two types of perforated spargers. Left: perforated plate, Right: perforated pipes
(Kulkarni and Joshi, 2011)
While various materials have been employed to manufacture porous spargers for
industry, such as glass, ceramic, metals, and fabric, the most accepted type for gas
dispersion is the sintered porous metal sparger. At the present time, the Mott
Corporation, established in 1959, is the lead company in the manufacture of porous
16 |
Virginia Tech | spargers through the use of different metals and alloys. The principle of sintered porous
metal spargers is to introduce gas into the liquid/pulp through thousands of tiny pores,
creating far more numerous smaller bubbles than with drilled pipe spargers. Sintered
porous metal spargers result in a larger gas-liquid/slurry contact area thereby reducing
the time and volume required to disperse gas into liquid/slurry. The thousands of pores
over the surface allow a large volume of gas to be released with a high specific area
(Mott Corporation, 2015). Mott’s porous spargers are not only known for their uniform
gas dispersion, but they are also known for their rigid and durable construction.
Sintered metal spargers are comprised of powdered metal which has been
ligated together by subjection to heat below its melting point. This technique produces
average pore sizes in the 60 to 100 microns range, thereby allowing them to produce
extremely fine bubbles. The most common metals and alloys used in the construction of
porous sparger’s are aluminum, stainless steel, hastelloy, inconel, nickel, titanium, and
alloy 20. The choice depends on the application and special customer requirements
such as greater temperature and corrosion resistance (Mott Corporation, 2016). A study
conducted at the Aristotle University of Thessaloniki found that sintered metal spargers
with a smaller average pore diameter have a more uniform porosity and therefore
maintain a more even air distribution. Along with this, a study conducted by the
University of Florida found that the average diameter of a bubble emitted from a
sintered aluminum or stainless steel sparger ranges from 0.7 to 0.9 millimeters (Kazakis
et al., 2008).
Although sintered metal spargers generate very fine average bubble diameters,
they still experience plugging problems when exposed to slurry even at a low
17 |
Virginia Tech | percentage of solids concentration. For example, according to a study by Rosso and
Stenstrom (2006) conducted at 21 wastewater treatment facilities, sintered porous metal
spargers used at the facilities require periodic cleaning with water and acid to prevent a
rapid decline due to slimes plugging. The study showed that porous spargers require
filtered air and water to promote successful continuous flotation, both of which are
difficult to attain in mineral processing applications.
Two types of perforated spargers, namely single-phase and two-phase, can be
arranged in multiple configurations inside a vessel or tank. A single-phase sintered
metal sparger introduces air only through a porous membrane directly into the column,
whereas a two-phase sintered metal sparger injects air through porous media around
the circumference of a moving stream of water (El-Shall and Svoronos, 2001). These
two types of configurations are applied in deinking flotation, wastewater treatment, oil
and water separation, hydrogenation, ozonation, pH control, and others. Unfortunately,
due to their high maintenance requirements, the use of single- and two-phase spargers
is not applicable to mineral flotation processes and is typically confined only to
laboratory settings.
1.4.2.2 Dynamic Sparging Systems
1.4.2.2.1 Jetting Sparger System
As previously mentioned, one of the biggest drawbacks to the porous sparging
method is plugging when operated in the presence of solids. This prevents them from
being used for mineral processing in the mining industry. For this reason, several
groups including the U.S. Bureau of Mines (USBM), Cominco, and Canadian Process
Technologies (CPT) have developed various forms of high pressure “jetting” spargers
18 |
Virginia Tech | (Kohmuench et al., 2007). A jetting sparger is a device that allows numerous air bubbles
to emerge from a small circular orifice known as a nozzle. As Finch (1994) explains,
inside a sparger, “bubbles form as a result of instabilities of the jet surface,” and the
amount of bubbles and their sizes are dependent on the length of the jet (Finch, 1994).
A sparger device is meant to be operated at high pressure to generate bubbles, further
reducing the problem of plugging, which porous spargers present. However, the high-
pressure requirement translates into an increased horsepower demand, and, therefore,
increased operational cost.
Cominco and the US Bureau of Mines created the first two-phase, high velocity
spargers. These spargers mix water and high-pressure air through a small nozzle inside
the column before the discharge occurs. The purpose of adding water to the sparger is
to create a finer bubble distribution inside the column by shearing the incoming air
passing through the sparger (Finch, 1994). The addition of water to the sparger was first
proposed to be less than 1% of the volume of gas. Though the jetting spargers
proposed by USBM and Cominco showed great improvements in bubble distributions
and the possibility to be used for mineral processing, the inability to maintain them
without shutting down column operations was still unattainable. Later, a Canadian
company called Canadian Processing Technologies, Inc. (CPT) developed a single air
phase sparger, called SparJet, and introduced the concept of on-line maintenance.
SparJet is a removable air lance that ejects pressurized air from a single nozzle through
the side wall of the flotation column. The concept of on-line maintenance is achieved by
arranging multiple air lances of varying lengths around the column perimeter. Air is fed
into the end of each sparger via tee-valves, which allow the airflow to be adjusted or
19 |
Virginia Tech | completely blocked in the event of pressure loss or for required maintenance. The
arrangement of multiple spargers around the column not only allows for continuous
flotation operation, but it also ensures aeration of the full cross sectional area of the
column.
CPT made a number of improvements to previous design of their sparger, which
eventually lead to the development of the SlamJet sparger. To facilitate maintenance
and prevent slurry from entering the airline, CPT replaced the tee-valve system with a
high-tension spring located in the sparger’s cage. The spring controls the nozzle
aperture as pressurized gas enters the sparger, allowing the position of an internal rod
to move towards or away from the nozzle aperture. The spring tension can be adjusted
by loosening or tightening a screw located at the end of the sparger, as shown in Figure
1.5 (Kohmuench et al., 2012). With this design, if air pressure were lost, the spring
would close the nozzle of the sparger, preventing the backflow of slurry into the air
system.
Figure 1.5 SlamJet sparger with its main components
CPT’s new design also maximized bubble population by introducing Finch’s
concept into the operation of the sparger. According to the concept, the total population
of bubbles can be acquired by extending the jet length into the column. This can be
20 |
Virginia Tech | achieved through an increase in air density by adding water (Finch, 1994). With this in
mind, the SlamJet’s performance is enhanced with the addition of water below the air
supply manifold. High pressure-water and air enter the lance together and are
discharged into the column.
In 2007, the Eriez Flotation Division (EFD) acquired CPT. EFD claims that the
flotation kinetics in the column is greatly improved due to the high rate of gas dissolution
achieved by the SlamJet sparger (Eriez Flotation Division, 2016). Thousands of these
spargers have now been placed into commercial service in the minerals processing
industries.
1.4.2.2.2 Microcel
Column flotation has been acknowledged as one of the best technologies
available to separate fine particles of valuable minerals from their associated unwanted
matter. However, the process is less efficient when ultrafine particles in the slurry have
to be separated (Yoon, Luttrell, Adel and Mankosa, 1992). It was not until 1988 that
professors working at Virginia Tech’s Mining and Mineral Engineering department
developed a new flotation technology called MicrocelTM Column Flotation. The
objectives of the MicrocelTM system are (i) to create microbubbles, normally in the 50 to
400 microns range, without creating plugging problems, (ii) to generate microbubbles
using slurry instead of fresh water in order to minimize fresh water demand, and (iii) to
ensure the bubble generator can be maintained and repaired as required without
equipment shutdown (U.S. Patent No. US5397001 A).
To accomplish the objectives of the Microcel Column, microbubbles are
generated by pumping slurry from the lower part of the column and passing it through
21 |
Virginia Tech | parallel, in-line static mixers, which conduct the slurry back to the column at a larger
height from the slurry exit port (Kohmuench et al., 2009; U.S. Patent No. US5397001
A). The air required for bubble generation is injected at a high rate into every static
mixer at the front end. The slugs of gas formed at the entrance are then broken by the
shearing action of the blades, which in turn create microbubbles ranging from 0.1 to 0.4
mm in size (Lakshmanan, Roy and Ramachandran, 2015; Yoon et al., 1992). A
schematic of the static mixer is displayed in Figure 1.6.
Figure 1.6 Schematic of the Microcel Static Mixer (Yoon et al., 1992)
The principle of Microcel technology is based on the improvement of flotation
kinetics by combining pressurized air at a high intensity with small bubbles, which in
turn enhances the frequency of bubble-particle collisions and attachment (Kohmuench,
et al., 2009; Lakshmanan et al., 2015). Yoon describes the process taking place in the
MicrocelTM Column Flotation as a three-stage flotation circuit. The use of the static mixer
in column flotation applications are as follows: (i) a roughing stage, whereby air rising in
the flotation column collides and attaches with hydrophobic particles that are flowing
downward into the column flotation; (ii) a cleaning stage, whereby risen bubble-particles
22 |
Virginia Tech | to an improvement in zinc grade by 0.04% and sphalerite by 0.16%. As a result, the
final recovery improvement was 2000 tons more of higher grade zinc concentrate per
year (Pyecha, Lacouture, Sims, Hope and Stradling, 2006). A similar study was
conducted in Peru at the Antamina copper/zinc mine. Microcel technology was
employed at Antamina for the cleaning of copper and molybdenum. The process
showed a reduction of bubble size from 3.7 to 2.6 mm, representing a 6% increase in
copper recovery and a 20% increase in molybdenum (Lakshmanan et al., 2015).
1.4.2.2.3 CavTube
The CavTube is a sparging device that uses hydrodynamic cavitation to generate
tiny bubbles, also known as pico-bubbles, in the order of 102 microns in size (Concha
and Wasmund, 2013). Hydrodynamic cavitation occurs when the liquid pressure is
abruptly reduced below its vapor pressure by subjecting it to high flow velocity (Fan,
Tao, Honaker, and Luo, 2010). In essence, the CavTube system is a venturi tube
wherein the liquid passing through the conical convergent zone increases its velocity
due to the dramatic reduction in diameter. This diameter change can be observed in
Figure 1.8. The cavitation phenomena results from a pressure change in the liquid while
crossing the tube. The liquid has a higher pressure and lower velocity prior to entering
the throat than while crossing it. After it passes the throat, the pressure decreases and
the velocity increases. This sudden contraction and expansion results in the cavitation,
also known as nucleation phenomena (Wasmund and Bain, 2014; Zhou, Xu and Finch,
1993).
24 |
Virginia Tech | Figure 1.8 CavTube sparging system in a clear plastic model (courtesy of Eriez Flotation Division)
Hydrodynamic cavitation was introduced as a sparging system by inducing the
flotation pulp and compressed gas into the venture tube (CavTube) at high velocity. The
high velocity combined with the throat geometry generate cavitation in the pulp, which in
turn improves flotation due to the attachment of ultrafine particles to the ultrafine
bubbles (Zhou et al., 1993; Kohmuench et al., 2012). At the same time, pico-bubbles
promote flotation by boosting the attachment of larger bubbles. The pico-bubbles serve
as an auxiliary collector of particles (Kohmuench et al., 2009; Concha and Wasmund,
2013). This phenomenon is illustrated by Figure 1.9. The utilization of pico-bubbles in
the flotation process decreases the required dosage of collector and increases the
probability of bubble/particle attachment. It also decreases the probability of
bubble/particle detachment (Kohmuech et al., 2009; Zhou et al., 1993; Wasmund,
2013). These improvements translate into the possibility of floating ultrafine particles,
which was impossible with previous technologies.
25 |
Virginia Tech | 2. EVALUATION OF AIR-INJECTION SPARGERS
2.1 Introduction
Column flotation cells have become the most popular machine design for
industrial applications that require high purity concentrates. The superior metallurgical
performance of column cells can be largely attributed to their unique geometry which
readily accommodates the use of froth washing systems. The wash water minimizes the
non-selective entrainment of ultrafine gangue material that would otherwise be
hydraulically carried in the water reporting to the froth concentrate. The larger height-to-
diameter ratio of columns allows a deep froth to be maintained, which is essential to
achieve even water distribution. With this unique feature, column cells can provide
impressive levels of metallurgical performance closely approaching the ultimate
separation curve predicted by flotation release analysis (Kohmuench et al., 2007).
Another very important feature of column cells is the design of the gas sparging
system. One popular choice in the minerals processing industry is the Eriez SlamJet®
sparger. As shown in Figure 2.1, this type of sparger operates by passing compressed
gas through a small discharge nozzle. Fluid turbulence created by the exiting gas
disperses and distributes small bubbles into the flotation pulp. The sparger is equipped
with an internal moveable rod that is attached to a pressure diaphragm in the back
housing of the sparger. The internal rod automatically moves back/forward and
opens/closes the nozzle outlet when the compressed gas is switched on/off. This
patented design effectively eliminates the accidental backflow of flotation pulp into the
injection tube during shutdowns. The sparger can be operated as a gas-only injector or
27 |
Virginia Tech | slugs of gas and unwanted burping. This undesirable condition can cause issues for
operators such as low recoveries, decreased capacities, and inefficient energy usage.
2.2 Theory
In column flotation, gas dispersion properties play an essential role in mineral
recovery. Bubble diameter (d ), gas flow (Q), and gas holdup (Ɛ) are some of the
b
properties that govern column cell performance. However, in order to control and
determine these properties, the gas dispersion provided by the sparging system into the
column should first be determined. This requires an analysis of internal sparger design
and nozzle discharge coefficient.
The design of the gas sparging system is perhaps the single most important
factor in determining the effectiveness of gas dispersion in column flotation. For
injection-type spargers, the design typically involves a nozzle throat that discharges
compressed gas directly into the flotation pulp through a small diameter orifice. Ideally,
the operating pressure is set so that the nozzle operates under choked flow conditions.
The choked flow condition occurs when high pressure fluid (air/water) is forced to pass
through a restricted orifice (nozzle, hole, orifice, etc.) into a lower pressure zone. There,
the velocity eventually reaches a point where it is choked which is known as “critical
velocity”. At this point, as observed in Figure 2.2, the flow velocity reaches a plateau
that is independent of the pressure differential. This velocity is known as sonic velocity
and its creation is based on the law of mass conservation.
29 |
Virginia Tech | Equation [2] can be used to determine the mass of fluid passing through
restricted orifices (e. g. nozzle orifices and valves) when the velocity is choked.
√
( )
[2]
where:
C = discharge coefficient
A = nozzle/orifice area
= gas density
From this very well-known equation (Loomis, 1982), it can be inferred that:
(i) for choked flow of gases, mass flow rate is independent of downstream
pressure and depends only on temperature and pressure on the
upstream side of the restriction;
(ii) the equation mathematically implies that mass flow rate is proportional
to hole area and square root of pressure; and
(iii) the mass flow rate is only weakly dependent on gas temperature (via
density).
These three phenomena can be observed in Figure 2.3. These two plots show that the
gas velocity exiting the nozzle reaches Mach 1 (1129 ft/s) when the absolute pressure
ratio exceeds about 0.528. The volumetric air flow rate, however, increases past this
point in response to simple compression of the upstream flow and not to an increase in
exit velocity.
31 |
Virginia Tech | Figure 2. 4 Comparison of theoretical pressure vs flow curve for nozzle and actual curve obtained
in jetting systems
In jetting spargers, like the commercially available Eriez SlamJet spargers, a flow
restrictor (an integration of spring, rod, and tip) moves back as the cracking pressure is
approached. This is due to the compressed gas that creates a counter-pressure to the
spring and moves it back, as already illustrated in Figure 2.1. Without the presence of
the spring, jetting spargers will match the theoretical pressure-flow relationship, also
known as critical flow. Although the flow restrictor is beneficial in the way it avoids back
flow of slurry into the air line in plant operations, it is also responsible for unwanted
pressure drop in some sparger systems. It can be a detrimental due to higher energy
consumption required to reach the theoretical curve. When the flow restrictor is fully
open, it is possible to achieve the highest velocity. The best gas dispersion is expected
under this condition. Figure 2.5 is a representation of this theory.
33 |
Virginia Tech | To facilitate calculations and keep consistency in this study, equation [2] can also
be written as (Loomis, 1982):
[3]
where:
Q = gas flow (volumetric flow of fluid)
A = area of orifice
C = coefficient of flow
P = upstream total pressure
u
T = Upstream total temperature
To make use of equation [3], it is essential to determine the coefficient of flow,
which entirely relies on the internal configuration and roundedness of the sparger.
These coefficients depend on the nozzle design, but normally are in the 97% to 61%
range for simple phase fluid and smooth edge nozzles. When dealing with two phase
fluids, such as gas/water, a second coefficient has to be considered in the previous
equation. To avoid confusion, the gas coefficient flow can be denoted as C , and the
n
water coefficient flow as C . Thus, equation [3] can be written as (Loomis, 1982):
w
[4]
Because the work in this study is empirical, the gas coefficient flow was
determined from experimental data. In light of this, the gas flow obtained from multiple
tests representing a set of data were averaged and fitted to a model. Likewise, the
35 |
Virginia Tech | water coefficient had to be determined. For this purpose, an empirical model based on
pressure gauge, water addition, and a fitting coefficient, was employed. This coefficient
is described by equation [5].
[5]
( )
where:
K = fitting coefficient
1
P = pressure
W = water flow passing through the nozzle
2.3 Experimental
For the evaluation of commercially available SlamJet spargers, a pilot-scale
continuous and closed loop air/water test column was designed and constructed at
Virginia Tech-Mining Engineering laboratory (Figure 2.6). The configuration of the
apparatus for running the tests consisted of two major components: gas flow apparatus
and pilot-scale column flotation. Additionally, a data acquisition system was developed
and implemented into the circuit for performance monitoring.
Gas-liquid injection rate, frother addition, and inlet pressure are the crucial
factors in running sparging systems. This work focused on the study of these factors
with the goal of finding the optimum operation of a sparger to assist plant operators in
improving recoveries from their column cells. To facilitate this goal, the current study
was conducted to quantify changes in gas flow rate, gas holdup, and degas time
obtained for different running conditions, two different internal sparger designs, and
several different nozzle sizes.
36 |
Virginia Tech | evaluate the predictive capabilities of Eq. [3], tests were conducted at low and high inlet
pressures, ranging from 40 PSIG to 80 PSIG with 10 PSIG increments; different water
flow rates of 0, 0.15, 0.30 and 0.45 GPM; and two different frother concentrations of 0
PPM and10 PPM.
2.3.1 Gas Flow Apparatus
The left side of Figure 2.6 provides a schematic of the experimental gas flow
apparatus constructed to evaluate the proposed SlamJet sparger system. During
operation, compressed gas (air) was introduced to a pressure regulator set to hold a
constant pressure of 620 kPa (90 PSIG). An on-off valve was installed after the
pressure regulator to initiate/terminate the gas flow during a test run (Figure 2.7). The
compressed gas from the on-off valve was passed to a gas rotameter and 0-134 KPa
(0-150 PSI) pressure gauge assembly. The rotameter was equipped with a manual
control valve that allowed for precise control of the gas inlet pressure. A check valve
was installed after the rotameter to minimize problems associated with the back-flow of
pressurized water into the gas monitoring instrumentation. The gas flow from the check
valve was passed into a distribution manifold that was connected via a flexible hose to
the sparger unit. The distribution manifold was equipped with another 0-134 KPa (0-150
PSI) pressure gauge so that the inlet pressure to the sparger hose could be constantly
monitored.
When required, water was added to the distribution manifold after passing
through another water rotameter, control valve and pressure gauge assembly. The
injection water was pressurized using a high-pressure multi-stage pump. A by-pass loop
and pressure relief valve was used to ensure that the high-pressure pump did not
38 |
Virginia Tech | overflow from the test column, made it possible to rapidly recharge water displaced by
the gas held up in the system during different test runs.
Two external site tubes for hydrostatic pressure monitoring were installed along
the column height for the manual monitoring of gas holdup. The connection ports for the
upper and lower manometers tubes were located at distances of 67.5 and 98.5 inches,
respectively, from the top of the test column overflow level. In order to monitor gas
holdup into the column, readings were made by taking multiple photographs to the
manometer tubes for each condition. Then, the readings were averaged to obtain the
“mean holdup (%)” values (Figure 2.8b). The average fractional gas holdup in the test
column (i.e., between elevations and ) was calculated from the level of liquid in the
first manometer using the expression:
. [6]
For comparison, the average fractional gas holdup in the upper section of the
test column (i.e., between elevations to ) was also calculated using:
. [7]
Generally, the holdup values determined in the upper section were lower than
those in the lower section due to an increase in bubble size resulting from the lower
hydrostatic head as bubbles rise to the top of the column.
The total fractional gas holdup (Ɛ) can be calculated from:
[8]
where:
h -h = delta height in manometer levels
1 2
H -H = delta height in manometer mounts
1 2
40 |
Virginia Tech | For the pilot-scale test column used in the sparging evaluations, the gas holdup
can be estimated from a simple volume balance given by equation [9], as illustrated in
Figure 2.9.
[9]
where:
Q = volumetric gas flow rate
g
Q = volumetric water flow rate
w
U = bubble swarm hindered rise velocity
b
X = column cross-sectional area
To improve accuracy and more precisely determine the gas holdup in the test
column, the manometer tubes were replaced with four high-speed electronic pressure
transmitters that were connected to a data acquisition system (LabView). The
transmitters were installed at different heights to assess potential differences in bubble
size distributions in each section of the test column. A combination of scheme pressure
transmitters and data acquisition system allows monitoring of the pressure differences
by real time. Likewise, to monitor gas flow in real time, an electronic vortex flowmeter
was installed along the airline right after the head gas supplier and connected to the
data acquisition system (LabView). The data acquisition system was set up so 10
readings per second can be obtained from the pressure transmitters and vortex
flowmeter.
Figure 2.10 shows an example of data obtained from each pressure transmitter
in a single test. The gas sparging performance was monitored by means of dynamic
pressure transmitter readings taken after the gas was shut off, 60 seconds after running
42 |
Virginia Tech | Figure 2. 11 Data collection using pressure transmitters
Two types of spargers employing the same nozzle diameter: SLJ and SLJ-TAJ,
5 different inlet pressures ranging from 0 psig to 100 psig with 10 psig increments,
4 different water additions in GPM, and
0 PPM and 10 PPM frother dosage conditions.
In total, the amount of data collected employing the data acquisition system was from 80
tests. Here it is easy to see the importance on relying on a balanced model that can
facilitate the process of analyzing, comparing, and predicting the percentage of gas
holdup and gas flow for each test. Figure 2.12 shows a balance model obtained from a
test performed on the SlamJet with Turbo Air Jet (SLJ-TAJ) at a low inlet pressure,
intermediate water addition, and 0 PPM frother condition. From this model, it can be
observed that the delta holdup into the column was 2.08%, 5.4 SCFM of gas flow during
aeration, and 10.98 seconds for a complete column degas once the gas has been shut
off. Due to human error, the “Gas Off” time that represents the time when the gas was
shut off, had to be manually input for each test.
44 |
Virginia Tech | 2.4 Results
2.4.1 Effect of Inlet Pressure
As stated in previous sections, the design of the gas sparging system is perhaps
the single most important factor in determining the effectiveness of gas dispersion in
column flotation. For jetting-type spargers, such as the patented SlamJet® technology,
the design typically involves a nozzle throat that discharges compressed gas directly
into the flotation pulp through a small diameter orifice. Ideally, the operating pressure is
set so that the nozzle operates under choked flow conditions. As indicated previously in
the theory section, Equation [10] describes the flow of gas passing through a restricted
opening (e. g. nozzle, orifice, hole, etc.) when air only is employed for the sparger
operation and Equation [11] when both air and water are employed for their operation.
[10]
[11]
Where:
Q = gas flow
A = area of orifice
C = coefficient of flow (air only)
Cn = coefficient of flow for nozzle without water
Cw = coefficient of flow multiplier for nozzle with water
P = upstream total pressure
u
T = Upstream total temperature
47 |
Virginia Tech | Once the flow coefficients are known, Equation [11] can be used to create a
model that predicts gas dispersion into the column at specific condition of pressure
gauge, nozzle size, and temperature. [Note: Numerical values of the flow coefficients
are considered proprietary since sparger manufacturers spend large amounts of time,
effort, and funds to accurately determine these values for their particular nozzle
designs.] Interestingly, Equation [11] shows that gas flow rate varies linearly with inlet
pressure (i.e., increasing the pressure by 10% increases the gas flow by 10%). This
inherent characteristic is an attractive feature of injection-type spargers since it allows
for simple throttling and control logic. On the other hand, Equation [11] cannot be used
to conclude that a 10% increase in orifice area will result in a 10% increase in gas flow
rate since C is typically also a function of nozzle design/diameter. Therefore, the
selection of an appropriate nozzle size is generally best made in direct consultation with
the sparger manufacturer.
To offer guidance in choosing the most appropriate sparger size, sparger
manufacturers can also provide plant operators with power demand, in terms of inlet
pressure, to achieve a desired gas dispersion into their column flotation operation for
the chosen sparger size. In Figure 2.14, normalized graphs of Gas Flow vs. Inlet
Pressure are shown for two different sparger sizes (nozzle diameters). The data in
these plots have been normalized by dividing the flows and pressures by either the
maximum gas flow rate (Qmax) or maximum pressure (Pmax). In general, a larger
nozzle size demands more energy, in terms of inlet pressure, than a smaller nozzle.
However, it cannot be concluded that a bigger nozzle size provides better performance
in terms of gas dispersion. The same gas dispersion can be achieved with different
48 |
Virginia Tech | 2.4.2 Effects of Water Injection
The addition of water improves gas dispersion by creating a jetting flow of high-
velocity water droplets that efficiently transfer kinetic energy into the flotation pulp as
they exit the nozzle. The resultant energy dissipation generates turbulent eddies that
help to shear any undispersed pockets of gas into smaller bubbles. The downside of
water injection is that it also reduces the gas flow rate. Since the reduction in gas flow
with water addition varies for each type and size of nozzle, the manufacturer should
generally be consulted prior to attempting this improvement to ensure that the gas
compressor and ancillary equipment can handle the new flow and pressure demands.
In order to illustrate the effects of water addition on sparger performance, several
series of experiments were conducted using the test column apparatus. In these
experiments, a SlamJet sparger equipped with a B-size nozzle was evaluated without
water addition (i.e., gas only) and with the addition of three different water injection
rates. In each run, the gas holdup (fractional volume of gas to total volume of gas plus
liquid contained in the column) was monitored using the externally mounted
manometers (Figure 2.8b). The fractional gas holdup (Ɛ) was calculated experimentally
by employing Equation [12] and [13], shown in previous section:
ε= (h h ) / (H -H ) [12]
1 – 2 1 2
[13]
Based on theoretical expression given by Equation [13], gas holdup must increase as
the gas rate increases and bubble size decreases (i.e., rise velocity of the bubble
swarm decreases). Thus, for a given gas flow rate, a higher holdup would be associated
with smaller bubbles resulting from improved dispersion (Miskovic and Luttrell, 2012).
50 |
Virginia Tech | The experimental data from the water injection tests are plotted in Figure 2.15.
As expected, the gas holdup in the column increased in proportion to the normalized
gas rate (Q/Qmax) for all water addition rates evaluated. More importantly, the gas
holdup versus gas rate curves shift upward to higher values as the water injection rate
increased. When operating at 70% of the maximum gas flow rate tested, the holdup
increased from about 15% for the gas-only case to about 17.5% with the addition of a
low amount of injection water (which represented about 50% of the maximum water
injection rate recommended by the manufacturer. The holdup further increased to near
20% by further increasing the water injection rate to the “normal” level typically
recommended by the sparger manufacturer. The holdup increased by nearly 25% as
the flow increased from zero (gas only) to the maximum value tested. These data
provide strong evidence for the important role of water addition in attaining good gas
dispersion for gas injection type spargers.
Figure 2. 15 Effect of gas flow rate and water injection rate on gas holdup (10 PPM frother)
51 |
Virginia Tech | 2.4.3 Effects of Water-Gas Mixing
In order to work effectively, the pressurized water added to a gas injection
sparger must be well mixed into the gas prior to exiting the discharge nozzle. For well-
designed systems, this mixing is accomplished using internal networks of gas and water
pathways that are incorporated into the structural design of the sparger. The pathways
provide turbulence that is sufficient to completely homogenize the gas and water
mixture, but not so intense as to create unwanted pressure drops that would otherwise
adversely impact the sparger gas rate, dispersion performance and energy demand.
The design of the network of mixing pathways, which is proprietary to each
manufacturer of commercial spargers, is a key feature that is often not considered by
plant operators when purchasing a new gas sparging system or when replacing existing
gas spargers from different suppliers.
The importance of the design of the gas-water mixing network is illustrated by the
test data shown in Figure 2.16. In this case, experimental tests were carried out using
two types of spargers: the standard SlamJet (SLJ) and the SlamJet with Turbo Air Jet
(SLJ-TAJ). These series of tests were conducted at 10 PPM of frother using either a
“low” amount of injection water (i.e., 50% of the water rate recommended by the
manufacturer) or a “normal” level of injection water (i.e., 100% of the water rate
recommended by the manufacturer). At each water addition rate, two sets of tests were
conducted with and without the network of pathways (TAJ) required to achieve complete
mixing of gas and water prior to exiting the discharge nozzle.
52 |
Virginia Tech | Figure 2. 16 Effect of gas-water mixing pathways on gas holdup (10 PPM frother)
The data plotted in Figure 2.16 indicate that there was only a very light increase
in gas holdup when using the mixing pathways for the test runs that used a “low”
addition rate of injection water. This suggests that the range of gas velocities spanned
in this series of tests were already sufficient to ensure good mixing of water and gas
without the need for any source of additional turbulence. However, when pushed to the
higher “normal” water injection rates, the sparger equipped with the mixing pathways
provided a notably higher gas holdup compared to the otherwise identical counterpart
that was not equipped with the mixing pathways. For a gas flow representing about 60%
of the full range tested, the gas holdup improved from about 17% to over 20% via the
incorporation of the mixing pathways (TAJ) as part of the sparger design. Once again,
this data suggests that small changes to the basic design of sparger components can
have a dramatic impact on gas dispersion performance.
53 |
Virginia Tech | 2.4.4 Effects of Frother Addition Point
One of the most important factors in determining the overall performance of a
gas sparging system is the type and dosage of frother employed. Frothers are
surfactants that lower the surface tension of the flotation pulp so as to permit the
generation of small gas bubbles and promote the formation of a stable froth phase.
Chemicals used commercially as frothing agents include various types of aliphatic
alcohols, natural (pine) oils and cresylic acids (Laskowski, J. 1989). For columns,
stronger frothing agents such as polyglycolethers may also be used to accommodate
the large froth depths required for froth washing. When used, these stronger frothers
are typically used as mixtures with other types of frothing agents to minimize the buildup
of persistent downstream froth in launders, sumps and piping networks.
Figure 2. 17 Effect of frother addition point on flotation recovery for plant sites (a) and (b)
One often overlooked factor in frother use is the location of the injection point.
For example, Figure 2.17 shows the effect of injecting frother into different locations
54 |
Virginia Tech | around a column cell. In each case, the total dosage of frothing agent added by the
reagent pumps was held constant. Five different frother addition methods were
examined: (i) all frother added to the feed sump, (ii) all frother added to the sparger
water injection pump, (iii) frother split equally between the feed sump and sparger water
injection pump, (iv) all frother added to the wash water drip pan and (v) frother split
equally between the wash water drip pan and the sparger water injection pump.
The evaluation of frother injection point was conducted at two different plant
sites. For the first site, the in-plant test data (Figure 2.17a) shows that the best overall
recovery of floatable solids was obtained when the frother was equally split between the
feed sump and water injection pump. This provided nearly a 3-4 percentage point
increase in recovery compared to adding all the frother into either the feed sump or
sparger pump alone. However, for the second plant site, the test data plotted in Figure
2.17b) indicated that the best addition point was the feed sump. This addition point
provided a recovery that was 3-4 percentage points higher than when added to the
sparger pump and 2-3 percentage points higher when split between the feed sump and
sparger water pump. Another important observation from both these plots is that any
addition of frother to the wash water resulted in a substantial decline in recovery. This
result was not unexpected since most of the wash water reports to the froth product
launder, thereby reducing the amount of residual frother available to enter the flotation
pulp for small bubble creation/stabilization. These data suggest that plant operators
should carefully evaluate frother dosage levels and frother injection points to identify
optimal operating conditions during routine performance audits of gas sparger
installations.
55 |
Virginia Tech | 2.4.5 Pressure vs. Time Monitoring
In order to determine improvements in gas dispersion offered by a new sparger
type called SlamJet with Turbo Air Jet (SLJ-TAJ) and the effect of frother dosage in
mineral processing applications, experimental tests were carried out in a pilot-scale
flotation cell at different frother conditions with two types of spargers offered by Eriez
Flotation Division: SlamJet and SlamJet with Turbo Air Jet. Both spargers were
equipped with the same nozzle size of 4 mm for a fair comparison. As previously
explained, the TAJ sparger type is a modified SlamJet sparger that consists of multiple
blades designed and placed along the internal rod with the aim to mix air and water just
before the discharge takes place in the column cell.
The series of tests were conducted at 0 and 6 PPM of frother, at the same inlet
pressure, and using a “normal” level of injection water (i.e., 100% of the water rate
recommended by the manufacturer). At each frother dosage, two sets of tests were
conducted with and without the network of pathways (TAJ) required to achieve complete
mixing of gas and water prior to exiting the discharge nozzle. The tests were monitored
employing the pressure transmitter and data acquisition system (LabView) in a
continuous mode, but with different two time period: (i) the spargers run for a fixed time
(e.g., 180 sec) to allow air-holdup to come to steady state value and (ii) gas and water
flow rates completely cut off while the monitoring system actively records gas holdup as
a function of time (e. g., 120 sec) while gas releases from the column.
It is known that small bubbles have a lower rise velocity than large bubbles;
therefore, the longer the gas takes to be released from the column cell, the better the
gas distribution into the column due to the creation of smaller bubbles. The shape of
56 |
Virginia Tech | such a “degas” curve is shown in Figure 2.18. The data provided in this plot is indicative
of bubble size distributions present in the column that were generated by two different
sparger types at two different froth levels. From the profiles, it can be observed that the
sparger with the network pathways (Turbo Air Jet) improved gas dispersion when
operated under low froth condition (6 PPM) compared to the standard SlamJet. This
improvement can be observed from the longer period of time this curve takes to
completely release the gas and reach a stable condition of 0% gas holdup. From the
profiles it also can be observed that tests carried out with both spargers, standard
SlamJet and SlamJet with TAJ, at 0 PPM do not show any difference in terms of gas
dispersion. In conclusion, the TAJ modification appears to improve sparger gas
dispersion only when frother is added to stabilize the formed bubbles, but not under test
conditions with surfactant-free solutions.
Figure 2. 18 Effect of frother dosage for test performed with SLJ and SLJ-TAJ
57 |
Virginia Tech | 2.5 Discussion
2.5.1 Performance Modeling
To facilitate the selection of the sparger device, “Pressure vs Gas Flow” curves
were developed for all the available industrial-size SlamJet spargers manufactured by
the Eriez Flotation Division. These curves allow the plant operator to do a better
selection of the aeration system that can bring an optimal column flotation performance
without increasing energy consumption. It is important to know that this selection must
be made with a manufacturer representative since they have an ample understanding of
sparger performance, operation, and energy demand.
Figure 2.19 displays normalized graphs of Gas Flow vs Inlet Pressure for all full-
scale commercial operations of SlamJet spargers offered by the Eriez Flotation Division
for this study. The experimental data in these plots have been normalized by dividing
the flows and pressures by the maximum gas flow rate (Qmax) and maximum inlet
pressure (Pmax). As mentioned in a previous section, in terms of inlet pressure, a larger
sparger nozzle demands more energy consumption than a smaller sparger. However, a
larger sparger does not always deliver better gas dispersion to the column cell. Here it
is important to consult sparger manufacturers for a better selection of sparger that is
specifically designed for a given application. Furthermore, in order to compare all
spargers offered by Eriez Flotation Division, a plot of the experimentally measured and
mathematically predicted gas flow rates for the SlamJet spargers evaluated in this study
is provided in Figure 2.20. As already indicated, these data were collected using several
different sparger sizes that are used in full-scale commercial operations within the
minerals processing applications. Figure 2.20 shows that extremely good correlations
58 |
Virginia Tech | possible that the cracking pressure will be incorrectly set, resulting in spargers that
operate along different flow-pressure curves. This improper calibration of cracking
pressures results in less than optimal operating conditions since less gas flow is
obtained for the given compressor load. More importantly, the difference in gas flow rate
from each sparger has the potential to induce unwanted axial mixing of the flotation
pulp. The occurrence of such back-mixing has long been known to be detrimental to the
metallurgical performance of column-type flotation cells (Dobby and Finch, 1985).
Proper balancing of sparger flows and cracking pressures is necessary to ensure that
this condition never occurs.
2.5.2.2 Non – OEM Components
As with any engineered device, the performance of a gas sparger depends on
the integration of many individual components. For example, field experience has
shown that the replacement of wear-resistant ceramic nozzles with locally fabricated
metallic nozzles offers much shorter life spans and poorer long-term performance in
terms of gas dispersion due to nozzle erosion/corrosion. Such changes often have large
costly impacts on metallurgical performance while only saving pennies in replacement
costs. While operators often understand the importance of utilizing high-quality OEM
(original equipment manufacturer) parts for critical parts such as gas nozzles, they
occasionally fail to recognize that the replacement of other types of parts assumed to be
non-critical can also have a dramatic impact on sparger performance. For example,
Figure 2.21 shows a flow-pressure curve for a commercial sparger in which the gas
connector port was replaced with a similar connector. The data in this plot have been
normalized by dividing the flows and pressures by either the maximum gas flow rate
61 |
Virginia Tech | Investigation of Flash Flotation Technology Utilizing
Centrifugal Forces and Novel Sparging Methods
Dylan Mark Rowley
ABSTRACT
A new processing technique, centrifugal flotation, has been developed in recent research
projects to overcome the large residence times and fine particle limitations of traditional flotation
technologies. The major innovation in the area of centrifugal flotation is the Air Sparged
Hydrocyclone (ASH), which has proven capabilities in achieving quality products at specific
capacities greater than traditional flotation methods. However, the ASH technology ultimately
suffers from sparger plugging problems. Therefore, three unique flotation cyclone designs were
developed utilizing external sparging systems and control features to float fine coal. The
objective of each design was to create a system that mimics the behavior of the ASH technology,
while providing advantages in bubble generation and retention time requirements.
The evaluation of the three designs provided evidence towards the development of an
efficient centrifugal flotation technique. Evaluation of a flotation cyclone with an external
Cavitation Tube yielded a single-stage product with an ash content of 4.41% and a 45% recovery
rate in a retention time of 0.66 seconds. However, the system required 16 minutes to meet
comparable flotation yields and recoveries. The third design achieved a multiple-stage product of
11.32% ash at a 55% recovery in 20 minutes. These two designs provided low yield, high grade
products, but rejected a high percentage of hydrophobic particles and required high retention
times to meet typical flotation standards. In addition, these designs suffered by requiring high
frother concentrations and recovery could not be increased through increased aeration due to
design limitations. |
Virginia Tech | ACKNOWLEDGEMENTS
First and foremost, I would like to thank my advisor, Dr. Gerald Luttrell. His constant
guidance, technical, and operational knowledge proved beneficial to development of the research
project. His presence allowed me to grow professionally in the area of mineral processing. I
would also like to thank Dr. Greg Adel and Jaisen Kohmuench for serving as my committee
members.
This project was sponsored by Eriez Manufacturing, project number 441809, and I am
grateful for both their financial support and technical knowledge through the development of the
research.
I would like to extend thanks to Jim Waddell and Bob Bratton. I thank Jim for his
machining skills, knowledge, and tremendous patience. His experiences proved both valuable to
the project and the development of my personal skills. I thank Bob for answering the endless
questions I had and for his assistance in the construction phases of the research.
Finally, I would like thank my parents for giving me endless support on my journey
through both undergraduate and graduate studies. I will be forever grateful for their patience,
sacrifice, and support.
iii |
Virginia Tech | 1.0 INTRODUCTION
1.1 Background
1.1.1 Flotation Fundamentals
The processing of fine particles in mining and minerals industry has been a dynamic
challenge since the development of higher production extraction technologies and the continual
decrease of ore grades. Regarded as the most widely used separation technique in mineral
processing applications, flotation has developed into an efficient process to upgrade the fine size
fractions (minus 100 mesh) of raw ore at high capacities. Flotation exploits differences in surface
properties of particles through wettability. Particles that show an affinity for water are classified
as hydrophilic while particles that tend to repel water are considered hydrophobic. This
fundamental surface property is exploited to create an adequate bubble particle contact through
the use of conventional and column cells, chemical reagent addition, and novel bubble
generation designs to increase the flotation kinetics of hydrophobic particles and yield a quality
product.
Conventional and column flotation machines achieve the same objective of recovering
fine hydrophobic particles but varying in their methods and efficiency of separation (Figure 1.1).
Typically, a conventional machine consists of a large cell where the flotation feed enters the
lower portion and is mixed axially by a rotating impeller. The rotating stator draws both air and
slurry to disperse bubbles into the cell. In a column cell, the flotation feed enters near the top of
the cell and bubbles are generated through an external or internal sparging system (Kawatra,
2011). Other than the physical design differences, the column cell and conventional mechanical
cell contrast in their respective effectiveness of bubble and particle collection, particle and
bubble contact, and entrainment of fine gangue particles. In conventional cells, the bubble and
1 |
Virginia Tech | Figure 1.1. Comparison of Conventional and Column Cell (Kawatra, 2011)
particle collection zone is essentially the area surrounding the mechanical impeller where the
bubbles are originally dispersed. Column cells however use the entire volume of the cell to allow
for bubble and particle contact. The large collection area of the column cell also allows for the
separate entry points for the slurry and generated bubbles. The location of the bubble generation
point allows the bubbles to rise and directly collide with the suspended particles. In contrast, the
particle is dependent on the rotating impeller to provide adequate bubble particle contact, but the
action of the impeller often creates turbulence which can lead to particle detachment (Laskowski,
2001). Finally, column flotation is superior to conventional machines through the practice of
washing the froth with water to provide a counter-current flow in the froth phase and reduce
entrainment of ultrafine gangue particles reporting with the concentrate (Laskowski, 2001).
In either case, these bubbles ascend towards the froth phase colliding with both
hydrophobic and hydrophilic particles. Due to the surface properties of these particles, the
hydrophobic particles attach to the air bubble and travel towards the froth product while the
hydrophilic particles do not attach and report to the underflow. The ability of each flotation
2 |
Virginia Tech | machine to recover the hydrophobic particles is dependent of bubble particle attachment,
flotation reagents, and bubble generation which will be discussed in the proceeding sections.
1.1.2 Bubble Particle Attachment
The attachment between the bubble and particle in the case of flotation is the most
critical principal in the separation of hydrophobic minerals from hydrophilic particles. The
thermodynamics of a particle attaching to a gas bubble is shown in Figure 1.2. The
thermodynamic relationship between the particle and bubble can be described by Young’s
equation. The Young’s equation describes the interfacial energy equilibrium required for bubble
particle adhesion and relates the following interfaces: interfacial tension between liquid and
vapor ( ), interfacial tension between solid and liquid ( ), and interfacial tension between
LV SL
solid and vapor ( ). Using these energy relationships, particle bubble attachment will occur as
SV
the difference between the interfacial energies results in a negative value (Yoon, 2011).
[1.1]
Continuing the analysis and using the diagram shown in Figure 1.2, the contact angle θ
between the tension at the slurry vapor interface and the other interfacial tensions will be the
Figure 1.2. Thermodynamic Description of Particle Bu bble Attachment (Yoon, 2011). Yoon, R. (2011). Froth
Flotation: Thermodynamics of Flotation. Blacksburg, Virginia.
Used under fair use. Form attached.
3 |
Virginia Tech | main parameter in the particle adhering to the bubble.
[1.2]
Thus, in order to obtain particle adhesion, a contact angle of 90 degrees will be necessary to
reduce the difference between interfacial energies below zero (Yoon, 2011).
Since the development of flotation in the mineral processing industry, several models
have been studied to predict particle bubble attachment. In Sutherland’s model of particle
collection probability in “Kinetics of Flotation Process”, Sutherland concluded the recovery of
the particle is dependent on the probability of collision between a bubble and particle, the
probability for the particle to attach to the bubble, and the associated probability of the bubble
detaching from the bubble through the flotation process (Sutherland, 1948). Based on
Sutherland’s framework, works done by others have concluded the probability of particle
collision is a function of the ratio of particle diameter to bubble diameter (Yoon & Luttrell, The
Effect of Bubble Size on Fine Particle Flotation, 1989), and the effects of particle size on
recovery can be seen in Figure 1.3 (Gaudin, Grob, & Henderson, 1931).
Figure 1.3. Flotation Recovery versus Particle Size (Ga udin, Grob, & Henderson, 1931). The Effect of Particle
size on Flotation – AIME, Volume 414. Gaudin A., Grob J., Henderson H. Used with Permission from Steve
Kral, Editor of Mining Engineer Magazine at Society of Mining, Metallurgy, and Exploration.
4 |
Virginia Tech | Once the particle collides with the bubble, the particle has the opportunity to attach itself
to the bubble to allow for recovery and is known as the probability of attachment. After this
collision, the particle remains on the bubble surface for a period of time as the particles slide a
specific distance. The period of time that particle slides over the bubble is the sliding time and is
related to the velocity of the surrounding liquid as the bubble ascends towards the froth product
(Yoon & Luttrell, The Effect of Bubble Size on Fine Particle Flotation, 1989). Sutherland
detailed the necessary time for a particle to adhere to a bubble as the induction period
(Sutherland, 1948) thus defining a specific time period for the particle to slide along the bubble,
rupture the surface, and become attached. Yoon and Luttrell derived probability of attachment
equations for various flow conditions and concluded the critical parameters of attachment are
bubble radius, particle radius, bubble velocity, and induction time (Yoon & Luttrell, The Effect
of Bubble Size on Fine Particle Flotation, 1989). In validation of their conclusions, as induction
time decreases for a given bubble and particle size, probability of attachment increases and the
Figure 1.4. Analysis of Adhesion Probability in Flotatio n at Varying Induction Times (Yoon & Luttrell, 1989).
The Effect of Bubble size on Fine Particle Flotation - Mineral Processing and Extractive Review, Volume 5
Issue 1-4 PP 101-122, R. H. Yoon and G. H. Luttrell, Used with permission of Deborah East,
www.tandfonline.com, 2014.
5 |
Virginia Tech | same result occurs as particle diameter is decreased (Figure 1.4).
1.1.3 Flotation Reagents
The flotation process can be efficiently modified with the addition of chemicals which
alter the chemistry of the slurry or the mineral surface as most feed ores are not ideally suited for
the separation of valuable particles. The categories of flotation reagents include collectors,
frothers, and modifiers with each type serving a specific purpose to allow for optimum
separation. Collectors selectively increase the hydrophobicity of minerals by providing a thin
film of hydrocarbons over the selected surface through chemisorption or adsorption. This coating
of hydrocarbons essentially increases the contact angle of the particle and bubble aggregate thus
improving flotation. Collectors are typically added upstream of the flotation machines to allow
for proper conditioning. Although collectors do provide an added advantage of coating non
hydrophobic particles with a hydrocarbon film, some ores, like coal, require little to no collector
addition thus saving flotation costs (Kawatra, 2011).
Often utilized in junction with collectors are frothers which act as bubble and froth
stabilizers. These alcohol based or synthetic compounds reduce the surface tension of the liquid
and provide the necessary stability for air bubbles to remain in slurry, capture particles, and
ascend to the froth phase with the attached particles. Additionally, frothers create a stable froth
phrase allowing for efficient collection of the recovered minerals and ensuring particles do not
detached and descend back into the pulp phase of the flotation process (Kawatra, 2011). With
these characteristics of frothers, the ultimate effect of the addition of frother is an increase in
flotation rate thus increasing recovery. However, the increase in recovery leads to the increased
recovery of gangue particles (Klimpel, 1995).
6 |
Virginia Tech | Other flotation chemicals such as modifiers are added to the slurry in order to control the
selectively of the flotation process. These modifiers can serve multiple purposes as mineral
activators and depressants through controlling the way a collector adheres to each mineral
surface or regulating the pH of the pulp in order to induce flotation of different minerals.
Specifically, activators allow the adhesion of a specific collector to a mineral that would not
normally attach whereas depressants prevent the adhesion. The pH modifiers use acids and
alkalis to either lower and raise the pH of the pulp and create an effective environment for the
flotation of specific minerals (Kawatra, 2011).
1.1.4 Bubble Generation
Considering the critical necessity for bubble generation throughout the flotation process,
the principles of bubble generation in conventional and column cells have been constantly
analyzed. As described, slurry enters the impeller of a conventional cell, and air is either drawn
in by the vacuum created by the movement of slurry or provided by an installed blower. The
Figure 1 .5. WEMCO 1+1 Flotation Cell (WEMCO 1+1 Flotation Cell, 2010). WEMCO 1+1 Flotation Cell.
(2010). Retrieved from FL Smidth: http://www.flsmidth.com/~/media/PDF%20Files/Liquid-
Solid%20Separation/Flotation/Wemco11brochure.ashx. Used with permission from Andrew Cuthbert,
Director of Global Marketing at FL Smidth.
7 |
Virginia Tech | rotating impeller disperses the slurry air mixture and generates bubbles for particle bubble
attachment. Shown is a cross sectional view of the fluid and air motion in a mechanically
agitated cell (Figure 1.5).
Due to the turbulence created of the rotating impeller and the overall length, flotation
columns adopted sparging systems as the primary bubble generation system. Several sparging
systems have been developed since the air diffusers made from ceramic material in the early
applications of column flotation (Kawatra, 2011). Such systems include static or inline mixers,
porous tubes, and devices that utilize venturi principle (Figure 1.6). In the early developments of
the Microcel, Yoon et al. used a porous venturi-tube sparger where air was drawn into porous
tubing and bubbles were generated by the shearing force of the passing slurry (Yoon, Luttrell, &
Adel, 1990).
Canadian Process Technologies Inc. developed a sparging system that utilizes the
advantages of cavitation. An air and liquid mixture is subjected to a rapid decrease in pressure
Figure 1.6. Porous Venturi Sparging System (Yoon, Luttrell, & Adel, 1990) and CPT Cavitation Tube System
(Column Flotation Systems Cavitation Tube, 2009). Yoon, R., Luttrell, G., & Adel, G. (1990). Advanced Systems
for Producing Superclean Coal. Blacksburg: U.S. Department of Energy. Fair use as government
publication.Column Flotation Systems Cavitation Tube. (2009). Retrieved from Canadian Process Technologies
Inc.: http://efd.eriez.com/Products/Index/Cavitationtube Used with permission from Dr. Jaisen Kohmuench,
Deputy Managing Director at Eriez Flotation Division.
8 |
Virginia Tech | through a small orifice and cavitation occurs. With the presence of frother and the subsequent
increase in area, picobubbles are created for the flotation process (Column Flotation Systems
Cavitation Tube, 2009).
1.1.5 Flotation Kinetics
The flotation of particles can be described as a simple model which takes into
consideration the floatability of the particle and the time it is exposed to the flotation process.
The model shown below represents the recovery of a specific particle and is described as the
ratio of recovered mineral mass to the total recoverable mineral mass.
[1.3]
Where k is the flotation rate constant of the particle and t is the residence time (Adel, 2014).
The flotation constant of a mineral is dependent on physical particle parameters such as
particle size, hydrophobicity, mineral composition, and operating parameters such as aeration
rate and reagent addition. As previously described, the recovery has an optimum particle size
range and drastically lowers below 400 mesh and above 60 mesh. Thus, the flotation constant
can be controlled through efficient comminution and size classification of particles. In addition,
the recovery of a particle will be affected by the natural or chemically altered hydrophobicity and
consequently increase or decrease the flotation constant. The flotation rate of a particle is also
dependent on the mineral composition and whether the particle is a free pure mineral or consists
of a gangue mineral. This factor is least controllable as much of composition is dependent on the
ore particle size. Creating free particles can be achieved through regrind circuits but these
circuits decrease the particle size and can be cost intensive. An increase in operating parameters
such as aeration rate or frother concentration will lead to an increase recovery as the probability
of bubble particle attachment increases with these adjustments.
9 |
Virginia Tech | 1.2 Objectives
This project seeks to design, develop, and analyze new flotation techniques which take
advantage of centrifugal forces and novel sparging methods to achieve acceptable yield and
recovery rates while maximizing capacity per unit volume and shifting the particle recovery
curve towards finer size fractions. The basis for this design work is the introduction of
centrifugal forces to increase the flotation rate constant and the inertia for fine particles. For this
study, three separation designs are modified from Jan Miller’s Air Sparged Hydrocyclone.
Studies of the Air Sparged Hydrocyclone (ASH) have shown the ability to achieve comparable
yields and grades to typical conventional and column flotation units, but fundamentally have the
potential for plugging of the sparging system.
Three centrifugal flotation techniques were evaluated: flotation cyclone with fixed
pedestal dimensions and cavitation sparging system, flotation cyclone with variable pedestal
dimensions and cavitation sparging system, and flotation cyclone with variable pedestal
dimensions and tangential aeration sparging system. These designs potentially provide adequate
upgrades over not only traditional flotation units, but also improve innovative centrifugal
flotation techniques. Evaluating the three designs were based on determining equilibrium
operating and design parameters which created a satisfactory froth product. This research
evaluates the performance of the designs by concentrating coal from raw flotation feed.
1.3 Organization
This thesis is comprised of five major components describing the purpose of the research
and how it was performed. The previous introductory section details the fundamentals of
traditional flotation systems, the associated limitations, and how this research will potentially
overcome those limitations.
10 |
Virginia Tech | The literature review section presents the status of current technology in both the
conventional and column flotation and also the developments of centrifugal separation
technologies. The subsections of the literature review include the limitations of conventional and
column flotation, the advantages of centrifugal forces, and current centrifugal flotation
technologies. The conventional and column flotation section presents the current trends of
flotation cell size, required residence times, and fine particle flotation. The centrifugal forces
subsection details the mechanics behind liquid in centrifugal motion and how particles in this
field are subjected to specific forces which will aid in the flotation process. The centrifugal
flotation technology sections discuss the innovative developments that attempted to fill the gaps
of the traditional flotation practices with the aid of centrifugal motion.
The experimental section provides a detailed description of the samples, apparatus, and
testing procedures used in this research. The section contains relevant information about
operating and analysis equipment that was specifically used to evaluate the performance of the
flotation cyclone designs.
The fourth section provides results from the experimental testing of the flotation cyclone
designs and provides a discussion of the results. The discussion of the flotation cyclone considers
the comparability of the three designs to current flotation technologies and any advantages that
were discovered during the design testing.
The fifth and final section is an overview of the research project while providing
recommendations for future work of this project and the future research in the area of centrifugal
flotation technologies.
11 |
Virginia Tech | 2.0 LITERATURE REVIEW
2.1 Scope
The literature review covers three topics relevant to this project: flotation technologies in
current practices, centrifugal forces, and centrifugal flotation technologies. The first section
presents the limitations of the traditional flotation technologies used in today’s mineral
processing applications. The second section covers the fundamental mechanics behind
centrifugal forces since this is the main basis for the research. The third and final section
describes the development of current centrifugal flotation technologies and identifies the
limitations of those technologies.
2.2 Flotation Limitations
Although conventional and column flotation has developed into the most widely utilized
separation technology in the mineral processing industry, like other methods there are
operational limitations which can hinder efficiency. With respect to this project, one of the
limiting factors of flotation is the continual increase of cell volume and associated slurry
residence time. Described in the early flotation kinetics model, the recovery of a mineral with a
given flotation constant is reliant on the particle residence time. The longer a particle resides in
the active flotation collection zone, the higher probability the particle will be recovered. This
realization in flotation kinetics paired with the continual decrease in ore grades and particle size
has led to the cooperation between plant operators and manufacturers to install large volume
flotation cells.
12 |
Virginia Tech | Flotation feeds typically are between two and five percent solids by weight which results
in significantly large volumetric feed rates (Honaker, Kohmuench, & Luttrell, 2013). As the
production of minerals, metals, and coal continues to increase, an increase in cell volume will be
a likely consequence in order to satisfy the flotation feed standards thus maintaining efficiencies.
A plot created by Noble details the trend in flotation cell size over the past century (Figure 2.1).
As flotation cells started at volumes below one cubic meter, developments in the efficiency of
the flotation process has led to the exponential growth to volumes surpassing 100 cubic meters
and nearing 1000 cubic meters (Noble, 2013).
Several manufacturers offer large volume conventional and column flotation cells in
order to meet required slurry residence times and increased capacities. FL Smidth, a mineral
processing solutions company based in Utah, boasts the installation of 66 of their 250 cubic
meter Wemco Flotation Machines in Mexico and developed 350 cubic meter SuperCells for a
Figure 2.1. Trend of Flotation Size in Past Century (No ble, 2013). Noble, C. (2013). Analytical and Numerical
Techiques for the Optimal Design of Mineral Separation Circuits. Blacksburg: Virginia Tech. Used with
permission from author. Letter attached.
13 |
Virginia Tech | copper operation in the United States (Flotation Technology, 2010). Metso Minerals offers
flotation columns whose height is designed to give a slurry residence time between 10 and 20
minutes (High Recovery Flotation Column, 2014). In addition, Outotec’s TankCell flotation
machines allow 500 cubic meters of effective volume for the high recovery of fine particles
(Outotec, 2014). These flotation technologies offer benefits of high capacities, high efficiencies,
and low costs but ultimately suffer from the required long residence times to achieve quality
products.
The flotation process can be configured, as shown, as a time rate process in which
particle will be recovered at a specific rate based on size, composition, and hydrophobicity. In
correlation with the larger scale flotation cells, the retention times of the slurry is a critical factor
in the allowance of adequate time for the desirable particles to collide, attach, and rise with the
produced bubbles and is shown by the following equation:
[2.1]
Where t is residence time, V is flotation cell volume, and Q is the volumetric flow rate to the cell.
Luttrell et al. showed the effects of reducing retention times on recovery in a coal column
flotation cell (Figure 2.2). These three different feeds achieved approximately 80 percent
recoveries when the retention time was adequate, but when feed rate was increased, the recovery
fell significantly in comparison to its theoretical values (Luttrell, Kohmuench, Stanley, & Davis,
1999).
Additional studies have been conducted recently to determine the efficiency of multiple
stage column flotation and varying circuit configurations. Dennis Kennedy performed tests on
single stage and two stage column circuits in addition to comparing the recovery results of
changing a plant flotation circuit from a parallel circuit to a series circuit. Initial tests of the
14 |
Virginia Tech | Figure 2.2. Effects of Retention Time on Recovery (Lu ttrell, Kohmuench, Stanley, & Davis, 1999). Luttrell,
G., Kohmuench, J., Stanley, F., & Davis, V. (1999). Technical and Economic Considerations in the Design of
Column Flotation Circuits for the Coal Industry. SME Annual Meeting. Denver: Society for Mining,
Metallurgy, and Exploration. Used with permission Steve Kral, Editor, from Society of Mining, Metallurgy,
and Exploration. Letter attached.
differences between a single stage and two stage circuit, with respective retention times of 11.9
and 12.9 minutes, showed an increase from 74 percent recovery to nearly 80 percent when a two
stage circuit was implemented (Kennedy, 2008). These initial results led to the reconfiguring of
an in plant columns cell from a parallel to a series circuit. Instead of distributing the feed among
five columns, the feed was split to two columns and the tailings were reprocessed in the
remaining three columns. The change resulted in a recovery increase from 77 percent to
approximately 82% (Kennedy, 2008). Although these column circuit variations result in higher
recoveries and the same retention time as Kennedy shown, the increase recovery required an
increase volume. The increase in volume presents the overall issue with both column and
conventional flotation where high residence times and capacities are required to achieve desired
products.
15 |
Virginia Tech | Although not completely dependent flotation variables, the increase in flotation volume
can be attributed to the presence of finer size fractions in flotation feeds. Sutherland derived an
equation to adequately represent the rate of flotation that is dependent on particle size:
( ) [2.2]
where R is bubble size, r is particle size, λ is induction time, V is bubble velocity relative to
particles, N’ is the number of bubbles per unit volume of pulp, and θ is portion of particles
retained in the froth after fruitful collision (Sutherland, 1948). This characterization indicates that
a decrease in particle size will result in a decrease in flotation rate, which has been shown in
several experiments through the development of flotation technologies.
Figure 2.3 provides a representation of flotation rates that Fuerstenau collected from
several authors as the particle diameter varies (Fuerstenau, 1980). One significant reason for this
is the small mass of particles that lack the overall momentum to deviate from the fluid stream
lines surrounding the rising bubble and result in a collision needed for attachment.
Figure 2.3. The Effects of Varying Particle Diameter o n Flotation Rates (Fuerstenau, 1980). Fuerstenau, D.
(1980). Fine Particle Flotation. In P. Somasundara, Fine Particles Processing (p. 671). SME-AIME. Used with
permission from Jane Oliver, Manager of Book Publishing for Society of Mining, Metallurgy, and Exploration.
Letter attached.
16 |
Virginia Tech | applications, slurry is fed tangentially into what is known as a feed chamber or “header”. The
feed chamber consists of a vortex finder which serves as an outlet for overflow particles and a
mechanism to induce the centrifugal forces. This centrifugal force developed by the entering
slurry and assisted by the vortex finder is the driving factor behind accelerated sedimentation
resulting in the size or density separation of particles. An apex assembly located at the bottom of
the hydrocyclone provides an atmospheric connection and flow restriction to aid the swirling
slurry in constructing an air core and transporting material through the vortex finder (Wills &
Napier-Munn, 2006). These characteristics are the appealing factors of a flotation unit that
induces centrifugal forces and creates a zero pressure zone like a hydrocyclone.
To understand how these forces are developed and their associated impacts in the
flotation cyclone, an analysis can be performed by observing a simple particle moving with
liquid in a circular orbit (Figure 2.5). As shown a particle moving in a circular motion
experiences two forces exerted on it: a drag force by the liquid acting as resistance pulling the
particle towards the center of its motion and the centrifugal force which is moving the particle
away from the center of its motion. The force due to gravity will be neglected for simplicity of
Figure 2.5. Forces Developed in Particle in Circular Motion.
18 |
Virginia Tech | this explanation. In 1981, George Stokes developed an equation for settling velocity of a
spherical particle in fluid which can be seen below (Stokes, 1981):
⁄ [2.3]
where v is terminal velocity, g is acceleration, d is particle diameter, D and D are particle and
s f
fluid densities, and η is fluid viscosity. In the scenario of circular motion, the acceleration term
can be substituted by the centrifugal acceleration term given by the product of the particle
tangential velocity squared and the inverse of its radius of motion (Holdich, 2002)., i.e.:
⁄ [2.4]
Thus giving,
⁄ . [2.5]
The application of Stokes’ settling equation to a centrifugal motion scenario can have
added benefits to the design of a flotation cyclone. Stokes’ equation for a given application
where acceleration, particle diameter, and fluid viscosity is constant, the settling velocity relies
on the difference between the particle and fluid densities.
In addition to the application of Stokes’ settling equation to centrifugal motion, previous
work has been done specifically in the area of centrifugal separation in coal processing. Sokaski,
Sands, and McMorris applied the sedimentation forces of dense medium baths to the centrifugal
forces in cyclones (Sokaski, Sands, & McMorris, 1991). Sokaski et al. used the gravitational
force found in typical dense medium vessels which is used in the processing of coarse coal
particles shown in the following equation:
[2.6]
19 |
Virginia Tech | where d is the particle diameter, δ is the particle density, ρ is the liquid density, and g is the
gravitational acceleration. Substituting the centrifugal acceleration equation in for the
gravitational acceleration, g, the centrifugal force on a particle in a cyclone can be found
(Sokaski, Sands, & McMorris, 1991):
[2.7]
where v is the tangential velocity in the cyclone and r is the radius of the cyclone.
As previously described, flotation utilizes the ability of hydrophobic particles to attach to
air bubbles and ascend to the top of a flotation column due to the lower density than the
surrounding fluid. This fundamental principle of traditional flotation is the same principal used in
centrifugal flotation designs. However, an added advantage of centrifugal flotation over the
traditional flotation process is the presence of the centrifugal forces acting on the particles and
increasing the settling velocity. As shown by Stokes and Sokaski et al., an air bubble will be
subjected to a force pulling the bubble with a specific velocity towards the air core at the center
of the rotating motion. This is due to the bubble’s density relative to the surrounding liquid
which will potentially give the bubble particle aggregate a velocity towards the developed air
core.
2.3.2 Current Centrifugal Flotation Technologies
As the need for smaller yet equally efficient flotation increases in the mineral processing
industry, innovative technologies will continue to develop. Centrifugal flotation has been one
developing technology in order to overcome the shortcomings of traditional flotation by reducing
retention time and increasing flotation rate without sacrificing recovery. Each section provides
both improvements of the Air Sparged Hydrocyclone, Imhoflot G Cell, and TurboFlotation over
traditional flotation the associated limitations. In addition to the technology descriptions, any lab
20 |
Virginia Tech | scale or pilot plant tests performed using the mentioned technologies are provided with
procedures and results.
2.3.3 Air Sparged Hydrocyclone
Research towards developing the Air Sparged Hydrocyclone (ASH) began in the 1980’s
by Jan Miller and the Metallurgical Engineering Department at the University of Utah. The ASH
unit was designed considering both traditional hydrocyclone features while using a novel
sparging method in efforts to develop a high capacity, low volume fine particle flotation method
(Figure 2.6). A tangential feed and vortex finder configuration is located at the top of the unit
where slurry enters and develops a swirl flow. The swirl flow enters a porous cylinder which acts
as the sparging system where the swirl flow shears the entering air to create bubbles for the
Figure 2.6. Air Sparged Hydrocyclone Assembly (Ye , Gopalakrishnan, Pacquet, & Miller, 1988). Ye, Y.,
Gopalakrishnan, S., Pacquet, E., & Miller, J. (1988). Development of the Air Sparged Hydrocyclone - A Swirl-
Flow Flotation Column. Column Flotation '88 - Proceedings of an International Symposium (p. 9). Denver:
SME. Used with permission from Jane Oliver, Manager of Book Publishing for Society of Mining, Metallurgy
and Exploration. Letter attached.
21 |
Virginia Tech | attachment of hydrophobic bubbles. The swirl flow continues to travel until exiting the unit
through an annular discharge at the bottom of the hydrocyclone with the rejected hydrophilic
particles (Ye, Gopalakrishnan, Pacquet, & Miller, 1988). The mechanics and features behind the
ASH unit will be discussed in length in subsequent sections.
The utilization of centrifugal forces aids in the flotation of ultrafine particles that are
characterized by low flotation rates. In the early developments of the ASH unit, Van Camp
studied the effects of force fields surrounding the flotation of a particle while Miller et al.
derived expressions to determine the critical particle size that exhibits insignificant inertia to
attach to a bubble. Van Camp concluded the relationship between flotation rates and force fields
are directly proportional. In traditional flotation columns, the force acting on bubbles and
particles is due to gravitational acceleration thus increasing their respective flotation rates by a
unity factor (Van Camp, 1981). Miller tested the effects of increasing force fields and the
resulting effects lowered the critical particle diameter with the required inertial momentum to
collide with rising bubbles (Miller, Kinneberg, & Van Camp, 1982). The work done by Miller et
al. and Van Camp provided the base theory for the Air Sparged Hydrocyclone unit by providing
evidence that centrifugal forces will increase the flotation rate constant and shift the effective
size range of the flotation process towards the finer size fractions.
To help induce centrifugal forces, the Air Sparged Hydrocyclone clearly resembles the
typical hydrocyclone in mineral processing applications and utilizes a similar header unit. The
traditional hydrocyclone header consists of two features: a tangential inlet and a vortex finder.
Both features help create the swirl flow needed in the ASH unit while the vortex finder serves a
dual purpose as the overflow discharge. The tangential inlet allows the slurry feed to enter the
Air Sparged Hydrocyclone and induce rotational flow at a radius typically equal to the
22 |
Virginia Tech | hydrocyclone radius. The vortex finder acts as a fixed point for the slurry to rotate about and
extends down into the unit to continue the rotational flow and prevent short circuiting of the feed
(Miller & Kinneberg, Fast Flotation with an Air Sparged Hydrocyclone, 1984).
As stated, the Air Sparged Hydrocyclone develops a swirl flow that shears air from a
porous cylinder in efforts to collide with hydrophobic particles and be recovered (Figure 2.7).
These particles that attach to the generated bubbles move radially towards to the center axis of
the hydrocyclone and develop a froth phase. The froth phase moves axially through the vortex
finder and exits the unit as product (Ye, Gopalakrishnan, Pacquet, & Miller, 1988).
The generation of the froth core is assisted to the presence of a zero pressure zone similar
to that of an air core in a typical hydrocyclone that is created through the restriction or
stabilization of the swirling slurry. Through the experiments of Ye et al., controlling the
characteristics of the froth core depended on the annular opening, overflow diameter, and the
presence of new hydrophobic minerals (Ye, Gopalakrishnan, Pacquet, & Miller, 1988). The
underflow area restricts the slurry flow inside the hydrocyclone thus forcing the layer towards
the zero pressure zone and out the overflow. In addition, emulating a traditional hydrocyclone,
Figure 2.7. Varying Phases of Air Sparged Hydrocyclo ne (Ye, Gopalakrishnan, Pacquet, & Miller, 1988). Ye,
Y., Gopalakrishnan, S., Pacquet, E., & Miller, J. (1988). Development of the Air Sparged Hydrocyclone - A
Swirl-Flow Flotation Column. Column Flotation '88 - Proceedings of an International Symposium (p. 9).
Denver: SME. Used with permission from Jane Oliver, Manager of Book Publishing for Society of Mining,
Metallurgy and Exploration. Letter attached.
23 |
Virginia Tech | an increase in overflow diameter also increases the size of the froth core and consequently
increasing the overflow rate. Lastly, in order to increase the amount of new hydrophobic
particles to the froth layer, increased reagent concentration or gas flow rate was necessary (Ye,
Gopalakrishnan, Pacquet, & Miller, 1988). The froth phase experiments concluded the benefits
of both flotation and centrifugal separation units as the presence of a zero pressure zone allows
the froth to build and stabilize while controlling such factors which are often found in typical
hydrocyclone units.
The sparging method of the Air Sparged Hydrocyclone is the stand out feature which
enables centrifugal flotation for the unit. The sparging cylinder is comprised of micrometer sized
pores to create air capillaries and allow the air to enter the hydrocyclone (Miller & Kinneberg,
Fast Flotation with an Air Sparged Hydrocyclone, 1984). These pores can vary in size as Miller
and Kinneberg performed tests using 340 and 630 micron pore sizes while Lelinski, Bokotko,
Hupka, and Miller performed tests with pore sizes ranging between 20 and 90 microns (Lelinski,
Bokotko, Hupka, & Miller, 1996). The air traveling through these capillaries is sheared by the
swirl flow of the unit and a distribution of bubbles varying in size is generated. These bubbles
generated travel towards the zero pressure zone in the unit due to their low density relative to the
slurry (Miller & Kinneberg, Fast Flotation with an Air Sparged Hydrocyclone, 1984). This travel
distance is the main probability for attachment of hydrophobic particles to the bubbles, and
varying certain operating and design parameter of the ASH unit can increase the attachment
probability by bubble concentration, stability, and size.
The critical parameters which make bubble generation possible have been significantly
studied throughout the development of the Air Sparged Hydrocyclone. The study of critical
parameters done by Lelinksi et al. in 1996 aimed to numerically analyze the effects of frother
24 |
Virginia Tech | concentration, slurry flow rate, and pore size on average bubble size and bubble size distribution
(Lelinski, Bokotko, Hupka, & Miller, 1996). A fine (20-40 micron), medium (40-60 micron), and
coarse (70-90 micron) porous tube were used in the experiment to study the effects of pore size
on bubble generation while shear force effects were examined at flow rates between 35 and 70
liters per minute. For a study of the surfactant effects, sodium dodecyl sulfate at concentrations
between 0 to 10-3 moles were tested. At a fine pore size and constant flow rate, the average
bubble diameter reduced from 1028 microns to 286 microns. At constant porous tube size and
surfactant concentration, increasing the flow rate thus increasing the shear force reduced the
average bubble diameter from 1028 microns to 755 microns. Lastly, by varying the porous tube
size while holding other factors constant, it can be seen the average bubble size reduces from
1028 to 838 microns (Lelinski, Bokotko, Hupka, & Miller, 1996). In addition, at high flow rates
and surfactant concentrations, the porous tubes were able to produce bubbles distributions with
an average size of 161 microns. The work done by Lelinski et al. showed the bubble generation
method in the Air Sparged Hydrocyclone unit was controlled by the same factors as the
traditional flotation unit and the additional control of the flow rate. In conclusion, the ASH unit
was able to achieve small bubble sizes which are necessary for the flotation of fine particles.
To compare the Air Sparged Hydrocyclone with traditional flotation technologies, a
considerable amount of pilot scale units with various ores have been tested to analyze the
feasibility of the unit. One of the earliest evaluations of the Air Sparged Hydrocyclone was done
by Miller and Van Camp in 1982 to separate a traditional water only cyclone coal feed (-28
mesh) with 22-25 percent ash content from Cerro Marmon Coal Group in Pennsylvania (Miller
& Van Camp, 1982). This coal feed was primarily comprised of the finer size fractions with 50
25 |
Virginia Tech | Table 2.1. Parameters for ASH Unit in Cerro Marmon Coal (Miller & Van Camp, 1982)
Air Sparged Hydrocyclone Parameters
Air Flow Rate 6.6 L/s
Frother Concentration 20 ppm
Feed Rate 4 L/s
Percent Solids 3 %
Cyclone Diameter 6 in
Porous Cylinder Length 29 in
to 70 percent of the particles finer than 400 mesh. The operating conditions and design
parameters of the ASH unit used in the coal testing is shown in Table 2.1.
Miller and Van Camp compared the results from testing the Air Sparged Hydrocyclone
with a batch flotation test using a conventional flotation cell. The conventional bench cell was
treated with 0.25 kg/ton of frother and product was collected for two minutes. In comparison of
product ash contents, the Air Sparged Hydrocyclone separated a product with 16 percent ash at a
75 percent yield rate at a 3.35 second retention time while the batch scale flotation produced a
15.5 percent ash product at a 67 percent yield rate (Miller & Van Camp, 1982). The comparison
proves the improvement of the ASH unit over the traditional conventional flotation cell as the
hydrocyclone obtained the same product quality at a higher yield and a lower retention time.
In addition to the comparison of the ASH unit and batch flotation, the effects of air flow
rate and porous cylinder length were evaluated. The separation efficiency was evaluated at
cylinder lengths of 16 and 29 inches and air flow rates of 200 and 400 lpm using 20 ppm frother,
3 percent solids, and 240 lpm feed rate. Increasing the air flow rate from 200 to 400 lpm
produced the same quality product at 16 percent ash while increasing the yield from 52 to 75
percent (Miller & Van Camp, 1982). The increased air flow rate acted in the fashion a typical
flotation unit would as the yield should typically increase. However, an increase in the product
26 |
Virginia Tech | ash is usually associated with this change in air flow rate but the ash remained unchanged as
proven. Like the higher flow rate, increasing the cell length from 16 inches to 29 inches
improved the product quality from 22.5 percent ash down to 16 percent ash while increasing the
yield by 16 percentage points. The higher quality product at an increased yield rate is likely due
to the increased retention time (Miller & Van Camp, 1982). Although the increasing the cell
length negates the objective of the ASH unit, the retention time in comparison with the two
minute conventional batch cell test was still significantly lower. However, the air flow rate for
the conventional cell was not listed but as described, a 400 liter per minute air flow rate was
required to achieve concentrate values which could be a disadvantage of the system.
The effectiveness of the ASH unit in the separation of the various size fractions was
evaluated and compared to other traditional flotation and separation technologies (Table 2.2). It
can be seen that the evaluation of the Air Sparged Hydrocyclone significantly held the advantage
over the ash removal capabilities throughout the various size fractions of water only cyclones
and single stage flotation.
Although testing the Air Sparged Hydrocyclone unit with coal slurry was a simple yet
effective evaluation, flotation is a separation technology with superior efficiencies throughout the
Table 2.2. Ash Removal Comparisons between Separation Technologies (Miller & Van Camp, 1982). Miller, J.,
& Van Camp, M. (1982). Fine Coal Flotation in Centrifugal Field With an Air Sparged Hydrocyclone. SME
Mining Engineering, 1575-1580. Used with permission from Steve Kral, Editor for Society of Mining,
Metallurgy, and Exploration. Letter attached.
Ash Removal (Percent)
Air Sparged Hydrocyclone Unit
Mesh Single Stage
WOC
Size Flotation
Illinois 6 Beaver Creek Lower Kittaning
28x100 60-65 50-60 77-86 84.2 70.4
100x200 40-45 40-45 59-68 70.2 39.6
200x325 15-18 40-45 44-62 77 42.5
-325 0-5 50-55 57-67 81 54
27 |
Virginia Tech | various mining industries. Continuous testing of the system has been performed with a variety of
ores to evaluate the versatility. In efforts to demonstrate the capabilities of the Air Sparged
Hydrocyclone, Miller et al. performed an evaluation with fine gold ore in sand obtained from the
Colorado River in Utah (Miller, Misra, & Gopalakrishnan, 1985). Miller et al. performed several
tests with the ASH unit with specific design and operating variables (Table 2.3) and compared
the results with a ten minute conventional batch flotation experiment.
The feed used in the testing was obtained from a gravity plant and was considered
tailings at minus 28 mesh with a majority of the feed in the minus 400 mesh size fraction. An
analysis on the feed proved the ore contained gold concentrations of 0.01 to 0.02 ounces per
metric ton. This feed was separated into a concentrate and reject stream using the Air Sparged
Hydrocyclone and batch conventional cell. Both the ASH unit and the conventional cell were
operated at the same operating parameters and the results were compared (Miller, Misra, &
Gopalakrishnan, 1985). It should be noted the air flow rate to the ASH unit was 100 lpm while
Table 2.3. Design and Operating Variables for Gold Flotation (Miller, Misra, & Gopalakrishnan, 1985). Miller, J.,
Misra, M., & Gopalakrishnan, S. (1985). Fine Gold Flotation From Colorado River Sand with the Air Sparged
Hydrocyclone. SME-AIME. Albuquerque: Society of Mining Engineers of AIME. Used with permission from Steve Kral,
Editor for Society of Mining, Metallurgy, and Exploration. Letter attached.
ASH Parameters
Design Variables
Cyclone Diameter 5 cm
Cyclone Length 52.5 cm
Pore Size 1 micron
Pedestal Diameter 4.25 cm
Operating Parameters
Promoter 0.05 g/kg
Collector 0.08 g/kg
Frother 0.1 g/kg
Air Flow Rate 100 slpm
Solids Feed Rate 1 tph
Percent Solids 16 %
28 |
Virginia Tech | the bench test was 6 lpm.
The results showed the superiority of the Air Sparged Hydrocyclone over the traditional
conventional flotation cell. In the testing of the Colorado River Sand containing fine gold
particles, the ASH unit achieved a concentrate grade of 2.17 ounces per metric ton at a recovery
of nearly 81 percent while the conventional batch cell only recovered 55 percent of the gold at a
1.075 opt grade (Miller, Misra, & Gopalakrishnan, 1985). Table 2.4 shows the complete
comparison between two separation processes abilities on the gold ore. The comparison of the
two separation methods proved the ASH unit recovered gold particles at a higher rate and higher
quality than the conventional flotation method. The improvement showed a shift in the grade
versus recovery curve towards obtaining the highest grade at the highest recovery rates possbile,
but the discrepancy between air flow rates may show a flaw of the ASH unit.
As the development of the Air Sparged Hydrocyclone continued through various mineral
industries on a laboratory scale, larger scale units were implemented into existing plants to
evaluate the true feasibility of the technology. One of the major efforts to install the ASH unit
into a processing facility occurred at a Florida phosphate operation where Miller cooperated with
the Florida Institute of Phosphate Research. In this partnership, Miller et al. analyzed the
operational feasibility of producing a high BPL (Bone Phosphate of Lime) by replacing a
Table 2.4. Comparison of Separation Results for Fine Gold Ore (Miller, Misra, & Gopalakrishnan, 1985).
Miller, J., Misra, M., & Gopalakrishnan, S. (1985). Fine Gold Flotation From Colorado River Sand with the Air
Sparged Hydrocyclone. SME-AIME. Albuquerque: Society of Mining Engineers of AIME. Used with permission
from Steve Kral, Editor for Society of Mining, Metallurgy, and Exploration. Letter attached.
Comparison of Flotation Methods on Gold Ore
Air Sparged Hydrocyclone Conventional Batch Flotation
Stre am Weight % Grade Recovery Weight % Grade Recovery
Concentrate 0.39 2.17 80.98 0.7 1.075 55.81
Tail 99.61 0.002 19.02 99.3 0.006 44.19
Feed 100 0.01 100 100 0.014 100
29 |
Virginia Tech | rougher cleaner circuit installation with a single stage six inch Air Sparged Hydrocyclone. The
results from the tests proved the ASH unit was able to achieve capacities 50 times the capacity of
traditional flotation techniques and produce a 66 percent BPL concentrate at a recovery of 91
percent (Miller, Wang, Yin, & Yongqiang, 2001).
Although the installation of the single stage ASH unit was efficient in producing a
quality BPL concentrate, the major concern of the Air Sparged Hydrocyclone resulted from the
phosphate testing. After operating continuously for 10 hours, the porous cylinder, the sparging
method for the unit, started plugging as crud formed on the inner wall. This plugging interrupted
the flow of air entering the system and ultimately lowered the recovery of the unit (Miller, Wang,
Yin, & Yongqiang, 2001). Significant studies were performed after the installation to test various
porous cylinders to fix the plugging problem. Plastic, ceramic, stainless steel, stainless steel wire
mesh, and hydrophilic plastic porous cylinders were examined and all materials suffered from
the plugging issue. The stainless steel tube’s permeability decreased to 57% of its original value
after 16 hours of operations and was the least impacted material by the crud (Miller, Wang, Yin,
& Yongqiang, 2001). This problem leads to the additional motivation of this research in order to
develop a sparging method that prevents plugging issues during continuous operation.
2.3.4 Imhoflot G Cell
The Imhoflot G Cell is a novel concept developed by Rainer Imhof at Maelgwyn Mineral
Services to recover ultrafine particles using centrifugal forces and minimal retention times
(Figure 2.8). The device exposes the flotation processes (aeration, bubble particle contact, and
froth pulp separation) into single operations and combines it into one unit. Slurry containing
ultrafine particles enters a downcomer unit where the slurry is aerated with multi-jet venturi
systems where the slurry shears the air and produces micrometer sized bubbles. The aerated
30 |
Virginia Tech | Figure 2.8. Pictorial Description of Imhoflot G Cell (Ba ttersby, Brown, & Imhof, 2003). Battersby, M., Brown,
J., & Imhof, R. (2003). The Imhoflot G-Cell - An Advanced Pneumatic Flotation Technology for the Recovery of
Coal Slurry From Impoundments. Cincinatti: Society for Mining, Metallurgy, and Exploration. Used with
permission from Jane Oliver, Manager of Book Publishing for Society of Mining, Metallurgy, and Exploration.
Letter attached.
slurry splits into multiple ports that enter the cell tangentially to create centrifugal accelerations
ranging between 10 and 30 m/s2. The centrifugal motion accelerates the froth and pulp separation
as the froth travels towards the center of the cell and exits axially while the primarily hydrophilic
pulp exits from the bottom of the cell. With this design, the G Cell is able to achieve retention
times between 25 and 30 seconds (Battersby, Brown, & Imhof, 2003).
Several implementations of the Imhoflot G Cell in the mineral processing industry proved
its efficiency over the previous flotation process. A three stage G Cell configuration was
installed over a conventional rougher cleaner circuit at a kaolin processing operation where
particle sizes of a few microns exist. With three 1.8 meter G-Cells at a capacity of 110 cubic
meters per hour, the circuit was able to achieve a kaolin concentrate at a 7 percent increase in
recovery but a 0.4 percent decrease in grade. The reported retention time for the three cells was
120 seconds versus the 14 minutes required for the previous rougher cleaner circuit (Imhof,
Fletcher, Vathavooran, Singh, & Adrian, 2007).
31 |
Virginia Tech | In further efforts to determine effectiveness, a fine coal impoundment was treated with
the Imhoflot G Cell. Fine coal impoundments are typically comprised of minus 325 mesh
particles due to the inabilities of traditional flotation techniques to recover that size fraction. A
two stage G cell flotation circuit yielded a 12 percent ash product from a 32 percent ash feed.
The two stage circuit produced a quality grade product at a yield of 70 percent from the original
feed of 40 tonnes per hour (Battersby, Brown, & Imhof, 2003). These applications of the
Imhoflot G Cell not only prove the technology’s performance against conventional flotation
techniques but prove the idea of centrifugal flotation can be just as effective. However, the G
Cell may suffer from the same flaws of conventional and column cells where multiple stages are
needed which increase the required volume of the circuit.
2.3.5 TurboFlotation
The TurboFlotation system, developed by the Commonwealth Scientific and Industrial
Research Organization (CSIRO), is a compact flotation technology that isolates the froth
separation from the bubble particle contact and bubble generation zones and utilizes centrifugal
forces to speed up the pulp froth separation. The system consists of several individual
components which helps optimize the unit flotation processes: jet ejector for bubble generation,
motionless mixer to induce bubble particle contact, and a centrifugal flotation cell to separate the
hydrophobic froth from the hydrophilic pulp (Figure 2.9). The jet ejector creates a low pressure
zone by increasing the liquid velocity thus drawing air into the slurry and creating bubbles. The
static inline mixer creates turbulent conditions to allow for the bubbles and particle to collide.
The slurry enters the separation cell tangentially to induce centrifugal forces and speed up the
flotation rate of the bubble particle aggregate (Ofori, Firth, & Howes, 2000).
32 |
Virginia Tech | Figure 2.9. TurboFlotation System (Ofori, Firth, & Howes, 2000). Assessment of the Controlling Factors
in TurboFlotation by Statistical Analysis – Coal Preparation – Volume 21 Issue 4 PP 355-382 – Authors – P.
K. Ofori, B. A. Firth & T. Howes. Used with permission of Deborah East, www.tandfonline.com, 2014.
The developments of the TurboFlotation system have shown the technology’s ability to
achieve 7-9% ash content products from a coal feed containing 23% ash (Ofori, Firth, & Howes,
2000). Initial efforts of testing the operating parameters in coal flotation, a two liter separation
cell was operated with a slurry feed rate from 10 to 16 liters per minute and achieved yield
values ranging from 53 to 80 percent. The low volume separation cell achieved separation
efficiencies as high as 64 percent at a retention time between 8 and 12 seconds (Ofori, Firth, &
Figure 2.10. Pilot Plant TurboFlotation Yields and Ash Content (Ofori, Firth, & Howes, 2000). Assessment of
the Controlling Factors in TurboFlotation by Statistical Analysis – Coal Preparation – Volume 21 Issue 4 PP
355-382 – Authors – P. K. Ofori, B. A. Firth & T. Howes. Used with permission of Deborah East,
www.tandfonline.com, 2014.
33 |
Virginia Tech | 3.0 EXPERIMENTAL
3.1 Coal Samples
In order to effectively test the designed prototypes for flotation capabilities, coal samples
were obtained from a currently operating coal plant. The natural hydrophobicity of coal makes is
an ideal testing material as it requires little modification in the flotation process. The sample used
in the current test program was provided from a facility located in southwestern Virginia. The
preparation plant produces both thermal and metallurgical coal while serving several different
mines in the area. The plant is equipped with two Microcel flotation columns that each treat run-
of-mine (ROM) minus 100 mesh feed. The minus 100 mesh coal was 48.30% ash and pre-treated
with collector at 40 ml/min. The feed release analysis and column data were provided by the
company in Tables 3.1 and 3.2.
Table 3.1. Release Analysis of the Coal Sample.
Individual Cumulative Float
Float Mass Ash Mass Ash Rec.
Product (%) (%) (%) (%) (%)
C1 9.01 6.95 9.01 6.95 16.21
C2 15.35 7.69 24.36 7.42 43.62
C3 16.24 9.32 40.6 8.18 72.1
C4 11.35 16.47 51.95 9.99 90.43
T2 2 65.66 53.95 12.06 91.76
T1 46.05 90.75 100 48.3 100
Table 3.2. In-Plant Performance of Column Cells.
% Solids % Ash
Feed 5.38 44.87
Conc 12.9 9.98
Tails 3.89 80.56
Yield % 50.57
Recovery % 82.57
Frother (ppm) 4.2
35 |
Virginia Tech | 3.2 Prototype Designs
In order to develop a system that incorporated centrifugal forces to aid in the flotation of
fine particles, three designs were created following the operating and physical parameters of the
Air Sparged Hydrocyclone, classifying cyclone, and flotation column. Although following
similar operating and physical parameters, each design consists of a unique sparging method to
provide bubbles for particle attachment. The subsequent descriptions will explain each of the
design theories with the expected advantages over the Air Sparged Hydrocyclone.
3.2.1 Initial Design: Fixed Pedestal with Tangential U/F
The primary design of the flotation cyclone consists of a tangential inlet and a discharge
ball valve with a pedestal of fixed diameter and height. The tangential inlet provides a feed port
for the slurry where centrifugal forces can be generated similar to a hydrocyclone. The discharge
ball valve provides a control for the slurry level inside the cyclone which can dictate the
associated water split. The pedestal provides a source of resistance in the cyclone in order to
develop an air core. In addition to acting as a source of resistance, the pedestal supports the air
core which is where bubbles containing fine coal particles would report.
According to J.D. Miller, the best separation efficiency for fine coal in the Air Sparged
Hydrocyclone occurred at an overflow area to underflow area ratio of 0.9 (Gopalakrishinan, Ye,
& Miller, 1991). Therefore, a pedestal diameter of 3.5 inches is calculated based on the
determined overflow and cyclone diameter. In addition, according to the Air Sparged
Hydrocyclone patent, the pedestal height should be at least fifty percent of the cylinder length
which results in a 12 inch pedestal height for the initial design (Miller, 1981). The vortex finder
diameter and length were based on design equations:
[3.1]
36 |
Virginia Tech | Table 3.3. Dimensions of Initial Flotation Cyclone Design.
Primary Design of Flotation Cyclone
Inlet Dia. 1 in
Cylinder Length 24 in
Cylinder Dia. 3.786 in
Discharge Dia. 1 in
VF Dia. 1.5 in
VF Length 2.2 in
Pedestal Dia. 3.5 in
Pedestal Length 12 in
[3.2]
where D is the diameter of the cylinder (Mular, 2003). In addition to the vortex finder
dimensions, the 1-inch inlet diameter is based on the equation which relates the cylinder
diameter to inlet area (Mular, 2003), i.e.:
[3.2]
The dimensions of the primary design are shown in Table 3.3. For reference, a schematic
illustration of the test unit is shown in Figures 3.1 and 3.2.
The main difference between the Air Sparged Hydrocyclone and the initial flotation
cyclone (Figure 3.1) is the sparging method. As stated, the sparged method in the ASH method is
a porous cylinder where air is sheared by the rotating slurry. However, due to plugging of the
porous cylinder, a two inch Cavitation Tube is placed before the inlet of the flotation cyclone.
The Cavitation Tube is a novel sparging device design by Eriez Manufacturing which creates
picobubbles by subjecting slurry to a constricted area. The slurry velocity increases to a point
where the decrease in slurry pressure induces cavitation and bubbles are created and stabilized by
the presence of frother (Column Flotation Systems Cavitation Tube, 2009). The Cavitation Tube
is expected to provide an external sparging source which would provide adequate bubble particle
attachment probability by generating picobubbles.
37 |
Virginia Tech | Figure 3.1. Initial Flotation Cyclone (Yan, 2013) Yan,
Figure 3.2. Initial Flotation Cyclone Design in
E. (2013). Initial Flotation Cyclone. Eriez Flotation
Operation.
Division. Used with permission from creator. Letter
attached.
3.2.2 Secondary Design: Adjustable Pedestal with Tangential and Axial U/F
The second flotation cyclone design incorporated not only potential advantages over the
Air Sparged Hydrocyclone but also over the initial design of a flotation cyclone with a pedestal
of fixed height and diameter. The sparging method remained the same with the Cavitation Tube
providing picobubbles, but the secondary design changed the discharge method to a two point
underflow system where slurry can exit tangentially or axially (Figure 3.3). The addition of the
axial underflow makes it potentially possible to control the level of the rotating slurry and to
install a pedestal whose height could be varied.
The adjustable pedestal provides an enhanced control factor to aid in the flotation of
particles. In the initial design, the volume of the flotation column developed inside the
hydrocyclone was limited by the fixed pedestal height and diameter. The fixed pedestal
dimensions potentially leads to bubbles bypassing the pedestal thus missing the collection zone.
The new design allows the pedestal to be adjusted to recover those potentially bypassing bubbles
and also limit water reporting to the overflow. Also, the pedestal diameter can be varied which
38 |
Virginia Tech | previously stated plugging issue with the porous membrane, a series of tangential ports, one
millimeter in width, are created in a 2-inch cylinder. The design of tangential ports allows the
slurry to travel along the cylinder wall with minimal impedance as it shears through the air
exiting the port. In addition, the tangential orientation of the ports will limit the potential for
particles flowing into the port and prevent plugging.
Designing the flotation cyclone with an involute feed provides many advantages over the
previous tangential feed design. The involute feed subjects the entering slurry to centrifugal
forces before the slurry transitions into the aeration cylinder. These forces will exhibit a certain
degree of separation between recoverable coal particles and gangue. The involute feed also limits
the turbulence caused by inlet junction. The limitation of turbulence will allow the floated
product to freely move into the vortex finder and out the flotation cyclone.
Although this design uses a different aeration and feed system, the third design still
incorporates aspects from the second design. Included in this design is the adjustable pedestal
where the height and diameter can be varied to fit the optimum conditions. The pedestal will
serve the purpose of creating resistance in the swirl flow in addition to supporting the froth
column generated. The tailings will be discharged through the axial underflow provided by the
circular opening.
The dimensions of the third flotation cyclone are based on the Air Sparged Hydrocyclone
used during phosphate testing by Jan Miller in partnership with the Florida Institute of Phosphate
Research (Table 3.5). The third design was comparably modeled after the cylinder diameter,
aeration length, and inlet area of the air sparged hydrocyclone. The pedestal and axial underflow
dimensions will vary between tests to create best operation, but will mainly be determined from
Miller’s design parameters for the Air Sparged Hydrocyclone.
42 |
Virginia Tech | Table 3.5. Flotation Cyclone with Tangential Aeration Cylinder.
Tertiary Design of Flotation Cyclone
Inlet Area* 0.5 in 2
Aeration Length* 13 in
Cylinder Dia.* 2 in
A U/F Dia. varies in
VF Dia. 1.25 in
VF Length 1.5 in
Pedestal Dia. varies in
Pedestal Height varies in
*Denotes dimensions of ASH in Phosphate Testing
In efforts to experiment with different sparging methods, the third flotation cyclone
design with aeration chamber was modified. Instead of utilizing compressed air in the aeration
chamber, an idea of filling the chamber with a slurry bubble mixture was developed. The
justification is the swirl flow will draw the bubbles through the tangential ports and the swirling
particles will collide with the entering bubbles. This opposes the original aeration chamber
theory by individualizing the flotation unit processes. Slurry was drawn out of the 240 gallon
sump using a ¾ horsepower pump, sent through a static mixer, and entered the aeration chamber
(Figure 3.6). The bubbles generated by the static mixer will fill the aeration chamber and be
Figure 3.6. Tertiary Flotation Cyclone Design with Static Mixer.
43 |
Virginia Tech | 3.3.3 Ash Determination
Determining the ash content of the feed, product, and reject streams was critical in
determining the performance of the three flotation cyclone designs. Provided at the research
facility was the LECO TGA701 Thermogravimetric Analyzer and it assisted in the determination
of the sample ash contents. The analyzer uses a controlled environment to determine composition
of the materials of interest. There are multiple software programs that analyze weight loss of a
material as temperature varies in the controlled environment. In the experiments of the flotation
cyclone designs, the analyzer has an ash analysis program that measures the weight loss of
combustible material in the samples as temperature increases (Figure 3.9).
3.3.4 Operational Controls and Measurements
In order to control and monitor the operating conditions for the experimental evaluation
of the flotation cyclone designs, several control devices were installed. The controls were used to
set desired levels of operating conditions such as slurry feed rate and air flow rate. Controlling
the slurry flow rate was performed by a PLC instrument connected to the centrifugal pump motor
(Figure 3.10). The controller regulated the speed of the motor which in turn controlled the
rotating speed of the pump. A more manual approach was taken to control the air flow rate. An
Figure 3.9. LECO Thermogravimetric Analyzer. Figure 3.10. PLC Motor Controller.
46 |
Virginia Tech | air regulator and ball valve combination was used to respectively control the pressure and
volume of air entering the circuit.
In addition to the control mechanisms, measurement apparatuses assisted by providing an
analog output of the operating condition. To measure the volumetric flow rate of slurry passing
through the network configuration, a one inch Maglite Flow Meter was installed in the piping
network (Figure 3.11). The Maglite Flow Meter sent the measured flow rate to an installed
analog output device which displayed the system flow rate (Figure 3.12). The output device was
configured to display readings up to 100 gallons per minute of slurry. As for the volumetric flow
rate of air being injected in the system, a simple floating flow meter was installed after the
regulator but before the ball valve. Multiple air flow meters were available to accommodate both
high (6 – 60 cfm) and low (0.4 – 4 cfm) flow rates.
3.3.5 Flotation Chemicals
In order to increase to flotation rate, chemical reagents were added to the slurry before
commencing the tests. For collector addition, kerosene, a typically used reagent in coal flotation,
was added to the metallurgical coal samples. The amount of collector added depended on the
solids content of the slurry but the normalized collector addition was sustained between 0.2 and 2
Figure 3.11. System Slurry Flowmeter. Figure 3.12. Slurry Flowrate Analog Output.
47 |
Virginia Tech | lbs. of kerosene to one ton of solids. For the frothing reagent, Aerofroth by the American
Cyanamid Company was used to generate bubbles and stabilize the created froth. The Aerofroth
is a mixture of aliphatic hydroxylated hydrocarbons and was added to slurry in the 240 gallon
sump at concentrations between 7 and 60 parts per million.
3.3.6 Miscellaneous Equipment
Throughout the experiments, miscellaneous equipment was used in junction for the
analysis of the flotation cyclone designs. The most important tools, electronic scales, were
mainly used for weighing wet and dry samples taken
from the three designs. The OHAUS Defender scale
(Figure 3.13) was used in the measuring of wet
samples especially collected samples that weighed
over four kilograms. A smaller scale was the primary
measurement method for dry samples gathered from
the flotation cyclone experiments.
Figure 3.13. OHAUS Defender 5000 Scale.
3.4 Testing Procedures
3.4.1 Water Only
Throughout this research, using water only provided a significant advantage at evaluating
the three flotation cyclone designs. The advantages of using water as a testing fluid included
visibility and ease of froth generation. Feeding water into the flotation system allowed
observation of the fluid mechanics of each design and how those mechanics are affected by
varying the operating and design parameters. In addition, the surface tension of water can
easily be reduced by the addition of a simple detergent in order to generate bubbles. These
48 |
Virginia Tech | simple yet effective characteristics of water allowed the optimization of the three flotation
cyclone designs.
To evaluate the designs of the flotation cyclone, each design was constructed with
specific pedestal dimensions, discharge characteristics, and operating conditions. The 240
gallon sump was filled with water to the appropriate level to prevent any pump operating
problems. Using the pump PLC controller, the speed of the pump motor was adjusted to the
desired frequency and consequently the desired water feed rate. Injecting the air required
adjusting the regulator and ball valve to the desired pressure and flow rate. At these conditions,
the pump was started and the action of the flotation cyclone design was evaluated. If any
modifications were needed to achieve the desired output, the operating or design parameters
were adjusted. Once satisfied, the simple detergent was added for froth generation.
Testing each cyclone design with water as the media served the purpose of monitoring
the effects of varying operating and design conditions. The effects of changing operating
conditions were measured by varying one of the parameters and measuring the response of the
products. The same procedure was performed for the design parameters. A single dimension
such as underflow diameter was varied while the other dimensions were held constant. The
masses of the overflow and underflows were collected in a five gallon bucket during a five
second interval and weighed. These values were recorded and were further analyzed with
Design Expert software to identify the critical operating and design parameters. The Design
Expert software uses statistical analysis to help develop prediction models for experiments. In
this case, the software was used to determine correlation of input and output parameters. Other
input controls and measured results are listed with a brief description (Table 3.6).
49 |
Virginia Tech | Table 3.6. Description of Input and Output Variables in Water Testing Evaluations.
Input Variables Symbol Description
Feed Rate Q GPM
s
Underflow Diameter D GPM
u
Pedestal Diameter D inch
p
Pedestal Height H % Cyclone Length
p
Air Flow Rate Q scfm
g
Frother Concentration C f ppm
Tangential Flow Rate Q GPM
t
Overflow Diameter D inch
of
Output Variable
Underflow Flow Rate Q GPM
u
Overflow Flow Rate Q GPM
o
Pressure Drop ΔP psi
Water Height H inches (air core)
w
In addition to identifying the appropriate operating and design conditions for the flotation
cyclone, an additional goal of testing with water was to verify the design acts in a similar manner
to the original Air Sparged Hydrocyclone and traditional flotation techniques. With respect to the
Air Sparged Hydrocyclone, each design should mainly be controlled by the feed rate and
underflow characteristics. In regards to both the flotation techniques and the ASH unit, the
operational performance of the designs should be related to the frother concentration and gas
flow rate. Concluding each design mimics the previous flotation work performed by achieving
the desired product will potentially prove that the design is an upgrade over existing
technologies.
3.4.2 Coal Flotation
Once the major parameters were identified using the water testing, each flotation cyclone
design was evaluated at the optimum operating point using the provided coal samples previously
described. As stated, using a coal sample as the floatable material decreased the complexity of
50 |
Virginia Tech | this procedure. The purpose of testing with the coal will provide yield and recovery rates that
will be the ultimate evaluation of the designs.
Prior to performing the evaluations of the designs, the coal slurries were sampled to
determine the feed characteristics such as percent solids and ash contents. A small sample was
collected using a simple syphon technique and a two liter bucket. The slurry was weighed to
obtain an initial slurry mass for future calculations. All the moisture in the sample was removed
using the vacuum pump and industrial oven to obtain a dry feed sample and the resulting dry
weight was measured and recorded. Using the slurry and dry sample masses, the percent solids
by weight was calculated and recorded. The sample was then stored until after all required
samples were collected, weighed, and dried for the ash analysis stage.
Once a sample of the collective feed was obtained, the slurry was placed in the 240
gallon sump. The flotation reagents, the collector and frother, were added to sump as well and
allowed to mix with slurry in order to properly disperse both chemicals. The amount of reagents
added was based on both traditional reagent dosages for flotation and the Air Sparged
Hydrocyclone. Using the optimum operating and design conditions determined by the water
testing, the feed rate, air flow rate, and design dimensions were set using the various
experimental controls, and the evaluation of the design commenced.
As the flotation cyclone came to a steady state, the overflow and underflow streams were
sampled during a five second interval using five gallon buckets. Collecting these samples
represented a single stage flotation cell which will be the basis of comparison. The product and
reject samples were weighed to record the slurry masses for the calculation of percent solids.
These samples went through the dewatering and drying process to prepare the samples for the
ash analysis step.
51 |
Virginia Tech | In addition to evaluating the flotation cyclone design as a single stage process, the
designs were evaluated as a continuous process where the reject streams were recirculated to
recover coal not floated during the first stage. This process would act similar in theory to
industrial flotation circuits that have multiple stages to recover as much valuable mineral as
possible. For this process, PVC pipe was attached to the overflow launder to direct the stream to
a separate collection point and the reject streams were allowed to enter the sump for
recirculation. To begin the process, the sample coal slurry was placed into the sump and the
optimum operating and design parameters were established. The pump was started and the
flotation cyclone was allowed to come to a steady state. The first samples of product and reject
streams were collected and the process continued until multiple concentrates were collected.
These samples were weighed, dewater, and dried in preparation for the ash analysis procedure.
The ash analysis procedures used the dry coal feed, product, and reject samples collected
during the design evaluations and found the ash content of each sample using the previously
described LECO TGA701 Thermogravimetric Analyzer. A representative sample of each
individual sample was placed in the analyzer which raised the temperature of the control
environment to remove any organic material. The analyzer reported the mass of inorganic
material remaining in the sample which represents the ash content of each sample. These values
for each flotation cyclone evaluation are reported in the Results and Discussion section of this
project.
52 |
Virginia Tech | 4.0 RESULTS AND DISCUSSION
4.1 Primary Design: Fixed Pedestal with Tangential U/F
4.1.1 Water Only
Initial water testing of the first flotation cyclone design intended to show the proof of
concept. Figure 4.1 depicts the initial test configuration of the flotation cyclone. As shown, an air
core was developed at the top of the pedestal. This air core is the basis of the flotation cyclone to
allow the bubble particle attachment to travel toward the developed zero pressure center.
Although the initial design achieved the desired air core result using centrifugal motion,
an inherent disadvantage was identified. When frother and air was added to the system, it could
be seen that bubbles travelling down the cyclone had the possibility of missing the collection
zone and reporting to the underflow. This disadvantage was attributed to the stationary pedestal
design of the flotation cyclone. The stationary pedestal design was rigid in the sense that it could
not be adapted to changing operating and design conditions. Recognizing this disadvantage led to
the secondary design with the adjustable pedestal.
Figure 4.1. Development of Air Core in Initial Flotation Cyclone.
53 |
Virginia Tech | 4.2 Secondary Design: Adjustable Pedestal with Tangential and Axial U/F
4.2.1 Water Only
Initial water testing of the secondary flotation cyclone design intended to identify the
critical design and operating parameters. The testing did not include the presence of air or frother
to assure proper evaluation of the fluid mechanics and system outputs. For the first procedure,
varying the design parameters along with cyclone feed rate was performed. The primary goal
was to obtain proper design conditions for the secondary design. However, a secondary goal was
declared in attempts to verify the previous ASH dimension testing and conclude the design acts
in a similar manner. The underflow and overflow flow rates, cyclone pressure drop, and water
height were the measured output parameters. The results from testing the input variables are
shown in Table 4.1.
Table 4.1. Secondary Flotation Cyclone Operating and Design Evaluation.
Input Variables Measured Results
D (in) D (in) H (%) Q (GPM) Q (GPM) Q (GPM) ΔP (psi) H (in)
uf p p s u o
1.25 - - 60 28.51 8.03 9.5 1.74
1.25 2 25 60 24.92 13.58 9 1.83
1.25 2 50 60 26.46 12.45 8.5 1.71
1.25 2 75 60 27.70 10.80 8.5 1.74
1.25 2 100 60 28.40 10.42 9 1.74
1.25 3.5 50 60 28.78 9.77 8 1.71
2 3.5 50 60 35.87 1.84 6 1.89
2.5 3.5 50 60 41.70 0.00 5 0.33
3 3.5 50 60 43.17 0.00 4 0.00
1.25 1.25 50 60 25.07 12.62 8 1.80
1.25 1.5 50 60 26.27 12.17 8 1.83
1.25 2 50 60 25.08 12.59 8.5 1.80
1.25 3.5 50 60 28.63 9.07 8 1.71
1.25 2 75 40 23.37 3.62 4 1.64
1.25 2 75 50 26.05 5.59 6 1.67
1.25 2 75 60 27.70 10.80 8.5 1.74
1.25 2 75 70 33.52 11.18 11 1.74
54 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.