a parametric study of froth stability and its effect

Ž .Int. J. Miner. Process. 59 2000 25–43www.elsevier.nlrlocaterijminpro

A parametric study of froth stability and its effecton column flotation of fine particles

D. Tao), G.H. Luttrell, R.-H. YoonCenter for Coal and Minerals Processing, Virginia Polytechnic Institute and State UniÕersity, Blacksburg,

VA 24061-0258, USA

Received 12 March 1999; received in revised form 7 July 1999; accepted 7 July 1999

Abstract

Laboratory flotation tests have been conducted to examine the effect of froth stability on thecolumn flotation of finely pulverized coal. It has been demonstrated that the upgrading of coal in aflotation column can be significantly improved when froth stability is properly controlled through

Ž . Ž .the manipulation of appropriate variables such as gas flow rate V , wash water flow rate V ,g w

froth height, wash water addition point, and feed solid concentration. Increases in wash water flowrate and gas flow rate promoted froth stability, resulting in higher combustible recovery, butpossibly higher ash recovery as well. The optimum rates were determined to be 2 and 0.3 cmrsfor V and V , respectively. The specific influence of wash water flow rate on water recovery wasg w

found to be closely related to how the frother dosage was maintained. Coal particles couldstabilize or destabilize the froth, depending on their size and concentration in the cell. Those iny100 mesh fraction destabilized froth at lower concentrations and stabilized it at higherconcentrations while micronized particles always showed froth-breaking power. The froth profilesof solid content and ash content were established at varying wash water flow rates and wash wateraddition points, from which it was concluded that froth cleaning occurred primarily at thepulp–froth interface and drainage above the wash water addition point. q 2000 Elsevier ScienceB.V. All rights reserved.

Keywords: coal; column flotation; entrainment; froth properties; surfactants

) Corresponding author. Present address: Department of Mining Engineering, 234E MMRB, University ofKentucky, Lexington, KY 40506-0107, USA. Tel.: q1-606-257-2953; Fax: q1-606-323-1962; E-mail:[email protected]

0301-7516r00r$ - see front matter q 2000 Elsevier Science B.V. All rights reserved.Ž .PII: S0301-7516 99 00033-2

( )D. Tao et al.r Int. J. Miner. Process. 59 2000 25–4326

1. Introduction

It has long been realized that the behavior of the froth phase is important inŽ .determining flotation performance. Tomlinson and Fleming 1965 and Feteris et al.

Ž .1987 showed that the flotation rate constant is directly proportional to the probabilitythat a particle survives the cleaning action of the froth zone and reports to the froth

Ž .product. Yianatos et al. 1988 reported that the separation efficiency of column flotationŽ .depends on froth depth. Bisshop and White 1976 found the strong dependence on the

froth residence time of the amount of drainage of hydrophilic particles from the froth. Itis generally accepted that a froth of proper stability is essential for the achievement ofgood grade and high recovery.

Ž .From the study of two-phase column froths, Finch et al. 1989 concluded that washwater can stabilize the froth providing bias rate J )0.1 cmrs. Work with severalb

Ž .minerals Klassen and Mokrousov, 1963; Moys, 1989 has shown that an increase in gasrate would result in more stable froth and higher water recovery. Engelbrecht and

Ž . Ž .Woodburn 1975 and Feteris et al. 1987 have demonstrated that froth stability isŽ .dependent on the height depth of the froth zone. Froth would eventually collapse with

increasing height due to liquid film thinning by drainage.Mineral particles have been reported to show pronounced effects on froth stability.

Ž .Szatkowski and Freyburger 1985 observed that fine quartz particles rendered bubblesto be more resistant to coalescence and promoted the production of the stable froth.

Ž .Livshits and Dudenkov 1965 believed that only coarse particles are able to act asbuffers between bubbles and prevent bubble coalescence, consequently strengthening the

Ž .stability of the froth. Klassen and Mokrousov 1963 reported that more hydrophobicŽ .particles had greater stabilizing effects on the froth. Johansson and Pugh 1992 showed

Ž .that particles of intermediate hydrophobicity contact angle uf658 would enhanceŽ .froth stability but more hydrophobic particles u)908 would destabilize the froth while

Ž .more hydrophilic particles u-408 would not influence the froth properties. There isapparent controversy on this subject and more work is required to elucidate thedifferences in effects of particles on froth behavior.

Relatively few studies have been conducted which examined the impact of frothbehavior on the performance of column flotation cells. The froth phase is generallyneglected in previous work on coal column flotation. This constitutes a major deficiencyin these studies because it is the froth phase that is primarily responsible for theimproved metallurgical performance of columns over conventional flotation cells.

In the present investigation, comprehensive laboratory flotation tests have beenconducted to provide a better understanding of the behavior of the froth phase in columnflotation cells. Parameters examined included gas flow rate, wash water flow rate, frothheight, wash water addition point, feed solid concentration, etc. The froth stability wasmeasured by the water recovery in the froth product. The profiles of solid content andash content in the froth were established at various operating conditions to betterunderstand the cleaning action.

( )D. Tao et al.r Int. J. Miner. Process. 59 2000 25–43 27

2. Experimental

Samples of run-of-mine high-sulfur bituminous Illinois No. 6 coal were acquiredfrom a coal preparation plant and used in the column flotation tests. The majorproperties of the coal sample are shown in Table 1.

The as-received samples were immediately crushed to y6 mm using a laboratory jawcrusher, split into representative lots of approximately 1.5 kg each, and then stored in afreezer at y208C to minimize oxidation. Prior to flotation, the samples were dry

Ž .pulverized in a laboratory hammermill to below 150 mm d s34 mm and diluted to50

10% solids, unless otherwise specified, in a conditioning sump. Micronization of thecoal sample, if needed, was achieved by wet grinding of pulverized coal at 30% solids in

Ž .a 13.3-cm diameter stirred ball mill Union Process using 3.2-mm diameter stainlesssteel balls. A 30-min grinding resulted in a mean product diameter of approximately 5mm at which nearly complete liberation of minerals from coal matrix was observed. Thecoal slurry pH was in the vicinity of 6.5. A kerosene addition of 1 kgrton of feed coalwas employed in all tests.

A microbubble flotation column having a diameter of 5 cm and height of 170 cm wasused in the test program. A detailed description of this apparatus has been provided

Ž .elsewhere Yoon et al., 1989 . In order to normalize the effect of column diameter, flowrates were expressed as superficial velocities defined as flow rates divided by thecross-sectional area. Unless otherwise noted, all tests were carried out at aeration rate of1.0 cmrs, wash water addition rate of 0.25 cmrs, and slurry feed rate of 0.17 cmrs.The slurry was fed to the column 60 cm below the froth overflow lip. Wash water wasadded in the middle of the froth zone, i.e., 25 cm below the froth overflow lip when thefroth zone is 50 cm high. The froth level was controlled primarily by adjusting thetailing discharge rate. Betz M-150, a polypropylene glycol methyl ether with molecularweight of approximately 400, was used as frother that was added directly to the bubble

Žgeneration circuit at a constant rate of 15 mlrmin corresponding frother concentration.in the pulp was ;20 ppm . The average bubble size was about 0.43 mm at 20 ppm

frother concentration, as determined from the air hold-up measurement using a differen-Ž .tial manometer Xu and Finch, 1989; Mankosa, 1990 . In the tests designed to determine

the ash and solid profiles within the column froth, a sampling lance was used to takesamples at various points along the column height.

Table 1Properties of the as-received Illinois No. 6 coal sample

Ž .Ash % 10.8Ž .Moisture % 10.3

Ž .Pyritic Sulfur % 1.16Ž .Organic Sulfur % 2.09

Ž .Total Sulfur % 3.25Ž .Volatile Matter % 37.0

Ž .Fixed Carbon % 50.8Ž .Contact Angle 8 48


3. Results and discussion

3.1. Column operating parameters and their effects on froth stability

( )3.1.1. Gas flow rate VgŽThe gas flow rate is of great significance in determining flotation rate constant Yoon

.et al., 1989 . It also plays an important role in the establishment of froth. As shown inFig. 1, water recovery increased considerably with the gas flow rate, indicating that thefroth was stabilized. There are two possible mechanisms for the observed increase inwater recovery with increasing V . The first is associated with more bubbles generated atg

higher gas rates. Under the assumption that the size distribution of bubbles is indepen-dent of gas flow rate and the thickness of liquid film of each bubble is constant, as

Ž .proposed by Engelbrecht and Woodburn 1975 , the recovery of water is expected to belinearly dependent on gas flow rate. This is confirmed in Fig. 1 by the linear correlationbetween water recovery and gas flow rate from 1.25 to 2.0 cmrs. The apparentnonlinear dependence above V s2.0 cmrs indicated the contribution from the secondg

mechanism, i.e., increased froth stability. The decrease in residence time of bubbles inŽthe froth diminished the coalescence of bubbles and drainage of liquid Bisshop and

.White, 1976 . It was observed in experiments that at gas flow rate ;3.0 cmrs, the gashold-ups in both zones were equal to each other and the interface between them

Fig. 1. The effect of gas flow rate on froth stability and flotation performance.


disappeared, which is in good agreement with the previous result obtained from aŽ .two-phase column froth Yianotos et al., 1985 . This is because higher gas flow rates

generated more bubbles in the pulp which carried more water into the froth, resulting ina decrease in gas hold-up in the froth zone but an increase in the collection zone.

Ž .Laplante et al. 1983a theoretically predicted and experimentally demonstrated thatin a flotation system without a froth phase, the flotation rate constant increased to amaximum and then decreased as gas flow rate was increased. This is due to the fact thatthe gas flow rate has two counteracting effects on the rate constant. On one hand, highergas flow rates generate more bubbles, increasing the flotation rate. On the other hand,higher gas rates increase bubble size, resulting in a decrease in the flotation rate. The gasflow rate at which these two factors are of equal effectiveness gives rise to themaximum flotation rate constant. In contrast with Laplante’s conclusion, Mehtrotra and

Ž . Ž .Kapur 1974 and Engelbrecht and Woodburn 1975 reported that in the presence of thefroth phase, flotation rate constant increases with gas flow rate over its full range, whichis in agreement with the results in Fig. 1 that shows combustible recovery increasedconsistently with increasing gas flow rate. It is believed that the increase in flotationrecovery was related to the increased froth stability which offset the effect of increasedbubble size.

Fig. 1 also shows that both water recovery and product ash increased significantlyŽ .with increase in gas flow rate when V )2.0 cmrs. Coincidentally, Finch et al. 1989g

recommended from a study of two-phase column froths that gas flow rate should be lessthan 2 cmrs. They found from the impulse and response tracer technique that at low gas

Ž .rates V -1.5 cmrs feed water concentration approaches zero at a froth height ofg

about 10 cm. At V )2 cmrs, feed water can penetrate 70–80 cm into the froth. As agŽ .result, relatively deep froths )1 m are necessary for a 10-m high industrial column

when high gas flow rates are used. Therefore, it appears that gas flow rate of about 2cmrs is appropriate for column flotation to achieve both high combustible recovery andlow product ash.

( )3.1.2. Wash water flow rate Vw

The characteristic that distinguishes flotation columns from conventional flotationcells is the addition of wash water to the froth. Previous studies have shown its

Žimportance in upgrading the froth product Luttrell et al., 1988; Finch and Dobby,. Ž .1990 . Finch and Dobby 1990 claimed that V could be as low as possible, providingw

the bias water rate is greater than zero. This conclusion, however, is not consistent withŽ .other studies Parekh et al., 1986; Luttrell et al., 1988 . Two series of tests were carried

out to investigate this discrepancy, with special attentions paid to the effect of washwater on the froth stability.

Fig. 2 shows the results of flotation tests conducted under the condition that thedosage of frother was constant at 0.5 kgrton of feed solids. The water recovery initiallydecreased rapidly with increasing wash water rate and then levelled off at ;0.2 cmrs.Although the total amount of water in the column was increased with increasing washwater rate, the amount of water carried into the froth product decreased consistentlyfrom 0.132 cmrs at V s0.06 cmrs to 0.038 cmrs at V s0.2 cmrs due to the sharpw w

decrease in water recovery. For V )0.2 cmrs, the amount of water reporting to thew


Fig. 2. The effect of wash water flow rate on froth stability and flotation performance with frother dosage keptconstant in lbrton of feed solids.

froth product increased with increasing V . The initial decrease in water recovery asw

well as water flow rate into the froth product could be attributed to the dilution offrother in the column. The subsequent leveling-off in water recovery, or increase inwater flow rate, indicates that the effect of wash water on preventing coalescence ofrising bubbles and film thinning predominated.

ŽThe bias rate of wash water defined as the net flow rate of wash water through the. Ž .stabilized froth can be estimated from Eq. 1 ,

V sV yV s 1yR V yR V , 1Ž . Ž .b w wp w w w wf

where V is bias rate, V wash water flow rate, V water flow rate to the froth product,b w wpŽV water flow rate in the feed, and R is water recovery defined as the fraction ofwf w

.water reporting to the froth product . It was determined that V was y0.072 cmrs atb

V s0.06 cmrs and 0.079 cmrs at V s0.12 cmrs, respectively. The zero bias ratew w

was achieved at V f0.10 cmrs. However, the product ash was minimized by V )0.2w b

cmrs or V )0.25 cmrs. Apparently, bias rate just above zero is not enough to obtainwŽ .the best product grade. In fact, Yianatos et al. 1987 showed from the study of

industrial columns that bias rate of 0.3–0.4 cmrs is required to prevent feed entrainmentinto the froth.

Combustible recovery decreased over the entire range of wash water rate. This mayresult from increasingly larger bubbles associated with lower frother concentration in the


collection zone caused by increased wash water flow rate. Since the objective of theaddition of wash water is to achieve maximum ash rejection with a minimum loss ofcombustible recovery, the most appropriate value of V should be around 0.30 cmrs orw

Ž18 cmrmin, which is in excellent agreement with the previous study Luttrell et al.,.1988 . Studies on the froth profile which will be discussed later confirmed this

conclusion.When the frother concentration in the pulp was kept constant at 20 ppm by adjusting

the frother addition to the column as wash water rate increased, the froth behavior andflotation performance were considerably different from Fig. 2, as shown in Fig. 3. Theproduct ash was almost independent of wash water flow rate, which is in coincidence

Ž .with the study on sulphide minerals Finch et al., 1989 . The combustible recoverychanged with wash water flow rate in a way very similar to that shown in Fig. 2 exceptthat the variation was less pronounced in this case. The change in water recovery in Fig.3 was relatively small compared to that exhibited in Fig. 2. The amount of waterentrained in the froth product was increased with increasing wash water flow rate,indicating that froth was increasingly stabilized. The bias rate was consistently positiveat all wash water flow rates ranging from 0.125 to 0.413 cmrs.

Results in Figs. 2 and 3 suggest that the inconsistence in literature on effects of washwater on separation performance may be associated with the difference in the wayfrother dosage was regulated. It can be speculated that under conditions where bubbles

Fig. 3. The effect of wash water flow rate on froth stability and flotation performance with frotherconcentration in pulp kept constant.


are not heavily loaded and gangue particles are pure and completely hydrophilic, frothcleaning action may be insensitive to wash water flow rate, provided that it generatespositive bias rate. However, in cases where significant quantities of gangue particles are

Žentrapped into the froth or middling particles are moderately floatable e.g., coal–mineral.composite particles , wash water flow rate or bias rate may have to be maintained at a

certain level in order to achieve optimum product grade. The y100 mesh coal sampleused in the present study contained about 40% middlings and as a result a significantpositive bias rate was required for effective cleaning.

3.1.3. Froth heightIt is known that for the effective operation of a flotation column, the pulp–froth

interface must be maintained at a certain level. Too high a level may result in a poorconcentrate grade and the reverse situation may give rise to a reduced recovery due todecreased residence time of particles in the pulp. Since typical industrial columns have acollection zone of 10 m in height, relatively small changes, e.g., 0.5 m, in the level willhave only a marginal effect on flotation recovery. This is especially true when columns

Žare operating near their maximum carrying capacity Espinosa-Gomez et al., 1988;.Ynchausti et al., 1988; Finch et al., 1989 . Level changes are most likely to affect the

product grade.Fig. 4 shows that there is a linear dependence of combustible recovery, ash recovery,

and water recovery on the froth height. The linear correlations and negative slopes are in

Fig. 4. The effect of froth height on froth stability and flotation performance.


Žagreement with the results obtained in previous studies with pyrite Engelbrecht and. Ž .Woodburn, 1975 and with galena Laplante et al., 1983b; Feteris et al., 1987 in

Ž .mechanic batch flotation cells. Bisshop and White 1976 suggested that the drainage ofhydrophilic particles from the froth back to the pulp is dependent on the froth residence

Ž .time which was found by Moys 1984 to be directly proportional to the froth height.Deeper froth provides longer time for liquid and particles to drain from bubble surfaceand drop back to the collection zone.

The difference in slopes of three lines in Fig. 4 is of particular significance since itdemonstrates the cleaning action in the froth. The recovery of ash-bearing particlesincluding composite ones decreased more significantly with increasing froth height thanthat of combustible material. This may be caused by the fact that more hydrophobicparticles will be preferentially attached to bubbles if particles are subjected to repeateddetachmentrreattachment events in the froth due to coalescence that makes less bubble

Žsurface available. In addition, decreased water recovery with increasing froth height as.shown in Fig. 4 reduces non-selective hydraulic entrainment, resulting in a lower

Ž .product ash. Laplante et al. 1983b suggested that the froth transfer coefficientcharacterizing the behavior of floating species in the froth decreased with increasing thefroth depth, which is consistent with the results shown in Fig. 4.

It should be pointed out that tests with bubbles larger than 1 mm in diameter showedonly marginal improvement in selectivity with increasing froth depth, which may beresponsible for the fact that thin froth is usually used in mechanical cells and some

Ž .columns Amelunxen et al., 1988; Kosick et al., 1988 in which bubbles are believed tobe 1–2 mm in diameter.

3.2. Effects of coal particles on froth stability

Effects of solid particles on froth stability were investigated as a function of solidconcentration in the feed at a constant feed flow rate. Fig. 5 shows results obtained with10-mm coal particles. Maximum solid concentration in the feed was 5% above whichfroth was almost destroyed. When solid concentration in the feed increased from 0.025%

Ž .to 5% solid concentration in the cell varied correspondingly from 0.1 to 20.0 grl , thewater recovery drastically decreased from 65% to 3.5%, suggesting that hydrophobicfine particles destabilized froth. The froth-destabilizing effect of fine hydrophobicparticles was also related to the consumption of frother in the cell due to its adsorptionon solids, evidenced by the increase in bubble size.

Data collected with y100 mesh coal samples are shown in Figs. 6 and 7. Com-bustible recovery and ash recovery at extremely low solid concentration in Fig. 6 werenot available due to insufficient quantity of particles in the feed. Apparently, coarse coalparticles destabilized the froth at relatively low solid concentrations. However, waterrecovery increased with increasing solid concentration in the feed from 5% to 25%. It isvery interesting to point out that over this range of solid concentration, bubbles becameincreasingly larger with increasing solid concentration, possibly due to the adsorption offrother on particles, which would have reduced water recovery. The results stronglysuggested that froth could be stabilized by coarse hydrophobic particles at relativelyhigh concentrations.


Fig. 5. The impact of 10-mm coal particles on froth stability, bubble size and flotation performance as afunction of feed solids concentration.

Additional flotation tests with coal particles in different size fractions were performedŽ .at a constant solid concentration. It was observed results not shown that particles

between 30 and 150 mm stabilized froth and those smaller than 30 mm destabilizedfroth. Smaller particles showed stronger froth-destabilizing ability.

Ž .The above results are in good agreement with Lovell’s conclusion Lovell, 1976derived from the work on phosphate flotation. According to the protruding particle

Ž .theory proposed by Hemmings 1981 , the condition that is required to produce adestructive compressive stress reaction in the liquid lamellae is,

t ucosu- -cos , 2Ž .

d 2

where u is contact angle, t lamellae thickness, and d is diameter of particle or floccule.The condition that is needed to produce a supportive tensile stress reaction is,

tu- -cosu . 3Ž .

d

However, if u)908, particles always tend to destabilize the froth. This theory suggeststhat small particles are more likely to destabilize the froth and big particles tend to holdthe condition for a stabilizing tensile stress. Increased froth stability at higher solid


Fig. 6. The impact of y100 mesh coal particles on froth stability and flotation performance with feed solidsconcentration from 0.025% to 5%.

concentration is generally attributed to increased surface viscosity of liquid films whichhinders film drainage.

Ž .Dippenaar 1982a established that a particle causes a film to rupture only when thetwo three-phase boundary lines are forced to migrate to the same point on the particleand the liquid film is thinned to half the diameter of the particle or less, which alsoindicates that smaller particles destabilize liquid films more readily than bigger ones.

Ž .The flotation tests of Dippenaar 1982b showed that 0.16 mg of 5-mm hydrophobicparticles had the same froth breaking power as 18.8 mg of 54.5-mm particles.

Ž .Johansson and Pugh 1992 reported that particles in 74–106 mm fraction generallyincreased froth stability. This effect reached maximum at relatively low hydrophobicityand further increase in hydrophobicity exhibited negligible influence. For particles in

Ž .26–44 mm fraction, those of moderate hydrophobicity contact angle uf50–658

showed substantial froth-stabilizing effect. Less hydrophobic particles did not haveinfluence on froth stability and more hydrophobic ones destabilized froth. Moudgil and

Ž .Gupta 1989 observed similar behavior with phosphate fines. It is believed that thepresence of smaller particles with moderate hydrophobicity at the interface of thePlateau border increased the rigidity and surface viscosity. More hydrophobic particlespenetrated the gasrliquid interface and the capillary pressure accelerated liquid drainage.


Fig. 7. The impact of y100 mesh coal particles on froth stability and flotation performance with feed solidsconcentration from 5% to 25%.

Particles with low hydrophobicity remained in bulk liquid and could not affect frothstability.

It is apparent from the above results and discussion that there are many factorsdetermining effects of solid particles on froth stability. They include particle size,hydrophobicity, concentration, etc. More importantly, these factors affect froth stabilityinteractively. Inconsistent observations and conclusions reported in literature may stemfrom different conditions various investigators employed in their studies.

3.3. Froth stability and particle entrainment

The recovery of hydrophilic particles in flotation is mainly caused by entrainmentwhich is related to the froth stability. The establishment of relationship between frothstability and particle entrainment is of great importance for a better understanding ofmechanisms for collection of different particles. It is also necessary for operation anddesign of columns to minimize the entrainment by optimizing operating parameters.

Various flotation tests were conducted in which gas flow rate, feed solid concentra-tion, and froth height were changed to acquire different froth stability and particleseparation while the other parameters are maintained constant. The combustible recoveryand ash recovery are shown in Fig. 8 as a function of water recovery. A lineardependence exists between ash recovery or product ash content and water recovery.


Fig. 8. The correlation between combustible recovery, ash recovery or product ash and water recovery.

However, the relationship is not in complete agreement with the results of EngelbrechtŽ .and Woodburn 1975 who demonstrated that not only was the recovery of hydrophilic

Ž .species silica directly proportional to the water recovery but also the line crossed theorigin. This difference may arise from the fact that they used the artificial mixture offinely ground silica and pyrite as feed in which silica was perfectly free of pyrite. In thepresent study coal was pulverized to y100 mesh and minerals were only incompletelyliberated from coal matrices.

The relationship between ash recovery and water recovery can be described by Eq.Ž .4 ,

R sF qe R , 4Ž .a a a w

where R is ash recovery, R water recovery, e slope of the line or entrainment factor,a w a

and F is the intercept of the extrapolated line on the ash recovery axis. The values of ea a

and F depend on properties of feed, e.g., size, and the chemical environment in theaŽflotation column. For well-liberated or ultrafine hydrophilic particles, F s0 En-a

.gelbrecht and Woodburn, 1975 . The e and F in Fig. 8 were determined to be 0.68 anda a

20, suggesting that a 20% recovery of ash was accomplished by true flotation ofcomposite particles.

Combustible recovery exhibited different and complex dependence on water recov-ery. Mathematical models describing their relationship that are available in literaturecannot be used to fit all the data satisfactorily. In an attempt to develop a new model,


Fig. 8 was replotted on a full log scale, as shown in Fig. 9. The curve can be best fittedin two distinct regions by the equations:

log 1yR s log aqb log 1yR , 5Ž . Ž . Ž .c w

or

bR s1ya 1yR , 6Ž . Ž .c w

where R is the combustible recovery and a and b are constants. When R s0,c w

R s1ya, which can be considered as the contribution from true flotation. In thec

region for R between 0 and 0.18, the best fitting values for a and b are 0.218 andw

4.683, respectively, which indicates that true flotation resulted in a combustible recoveryof 78.2%. For the region with R ranging from 0.18 to 0.50, the respective values of aw

and b were determined to be 0.1183 and 1.446. The correlation coefficients are 0.938and 0.934, respectively, in these two regions. Apparently, true flotation is the predomi-nant recovery mechanism for combustible material. It should be noted, however, that therelative contribution of true flotation and entrainment to the overall recovery ofhydrophobic particles is strongly dependent on the particle size. For example, WarrenŽ .1985 showed that for iron sulfide particles of intermediate size entrainment played aminor role in overall recovery and true flotation was predominant, which is in agreement

Fig. 9. The relationship of combustible recovery and water recovery on the full log scale.


with the present study; for flotation of ultrafine particles in mechanical cells the overallrecovery was largely caused by entrainment and contribution from true flotation is small.

3.4. Froth profile

Particles that are collected by bubbles in the pulp and carried into the froth phase maydrop back into the pulp because of the continuous drainage of liquid and bubblecoalescence. Hydrophilic or relatively less hydrophobic particles are more likely todetach from the surface of bubbles, leading to the cleaning action in the froth.Consequently, the grade of particles in the froth could be a function of froth height. Abetter understanding of froth cleaning mechanism can be accomplished through theestablishment of froth profiles.

3.4.1. Effects of wash water flow rateFig. 10 shows the ash profiles in the froth at different superficial wash water rates. As

shown, an abrupt decrease in the product ash content occurs just above the pulp–frothinterface. This upgrading can be largely attributed to the reduction in percent areaoccupied by pulp and efficient cleaning action of the wash water at the interface whichprevents the ash-bearing minerals from entering the base of the froth. At low wash water

Fig. 10. The froth ash profile at different superficial wash water flow rates.


addition rates, ash content decreases slowly with height up to the wash water inlet. Thissection of the froth column is referred to as the stabilized froth zone since the flow ofwash water through this zone retards bubble coalescence and stabilizes the froth.Bubbles in this zone are relatively uniform in size and coalesce very slowly. Above thewash water inlet, the ash content again decreases remarkably with height. This section ofthe froth, which is called the draining froth, is similar to the froth typically found inconventional flotation cells. The cleaning action in this zone is a result of bubblecoalescence and consequent drop-back of some of entrained particles into the pulp.

The ash profiles obtained at higher wash water rates remain relatively constant fromthe pulp–froth interface to the top of the froth, which is in agreement with the

Ž .observations of Szatkowski 1987 . This is because hydraulic entrainment is largelyeliminated at the pulp–froth interface and, therefore, no significant additional upgradingwould occur in the draining froth. These results indicate that superficial wash water ratesgreater than ;0.25 cmrs are sufficient to prevent entrainment of feed water into thefroth product under the given experimental conditions. This value agrees very well withthe result discussed in Section 3.1.2.

3.4.2. Effect of wash water addition pointThe results shown above suggest that the wash water addition point may be a key

variable in column flotation. In order to examine this possibility, additional tests were

Fig. 11. The impact of the wash water addition point on the product ash content and the product mass rate.


conducted in which the wash water addition point was varied within the froth height.The aeration rate, feed rate and wash water rate were held constant throughout thesetests. The results are shown in Fig. 11. The ash content decreased as the wash wateraddition point was moved from the top of the froth down into the cell. The decrease inash content can be largely attributed to an improvement in froth washing associated withthe increased bias factor, which was 0.62 and 0.96 when wash water was added at thetop and at the bottom of froth, respectively. However, as the addition point was movedcloser to the pulp–froth interface, the stabilized froth zone became too short toeffectively prevent the entrainment of fine particles. As a result, a slight increase inproduct ash content was observed when the wash water addition point was moved tooclose to the pulp–froth interface.

Fig. 11 also indicates that improvement in product grade comes at the expense ofcolumn capacity. As shown, a lower wash water addition point produces a sharpdecrease in the mass rate of product reporting to the froth launder. The lower capacity isdue to the decrease in the froth stability in the froth drainage zone which occurs as thewash water addition point approaches the pulp–froth interface. Since wash water isineffective when added at the interface, the frother is not washed from the froth phaseback to the pulp, generating more stable froth. The increased froth stability leads to ahigher mass rate.

4. Conclusions

1. Increase in gas flow rate stabilizes froth and improves combustible recovery.However, gas flow rates higher than 2 cmrs result in sharp increase in ash recovery dueto increased nonselective hydraulic entrainment.

2. Effects of wash water rate on froth stability and flotation performance depend onwhether frother concentration in the column is maintained constant as wash water rate

Ž .changes. A significant positive bias rate )0.2 cmrs is required to eliminate hydraulicentrainment for the coal sample investigated. The appropriate wash water rate dependson aeration rate, frother concentration, coal characteristics, etc.

3. Micronized Illinois No. 6 coal particles have strong froth-destabilizing effect.Ž .Relatively coarse coal particles y100 mesh destabilize froth at lower solids concentra-

Ž .tions -20 grl in the pulp and stabilize froth at higher concentrations.4. Ash recovery is linearly dependent on water recovery while combustible recovery

shows a nonlinear dependence. A new model has been developed for the correlationbetween combustible recovery and water recovery.

5. At higher wash water addition rates, upgrading in a flotation column occursprimarily at the pulp–froth interface; at lower wash water rates, additional upgradingtakes place in the drainage zone. Froth phase should be made as deep as possiblewithout causing a significant loss of combustible recovery.

6. For a fixed set of experimental conditions, lowering of the wash water additionpoint into the froth increases bias rate, which may improve the effectiveness of the washwater and increase the product grade. However, excessive drainage of froth associatedwith lower wash water addition position will lead to an unstable froth which adverselyaffects column performance, particularly in terms of capacity.


Acknowledgements

The authors would like to acknowledge the financial support of the United StatesŽ .Department of Energy Contract No. DE-AC22-86PC91221 which made this work

possible.

References

Amelunxen, R., Lierena, R., Dunstan, P., Huls, B., 1988. Mechanics of column flotation operation. In: Sastry,Ž .K.V.S. Ed. , Column Flotation ’88. SME Annual Meeting, Phoenix, AR, pp. 149–156.

Bisshop, J.P., White, M.E., 1976. Study of particle entrainment in flotation froths. Trans. Inst. Miner. Metall.,Sect. C, Miner. Process. Extr. Metall., C191–194.

Dippenaar, A., 1982a. The destabilization of froth by solids: I. The mechanism of film rupture. Int. J. Miner.Process. 9, 1–14.

Dippenaar, A., 1982b. The destabilization of froth by solids: II. The rate determining step. Int. J. Miner. Proc.9, 15–27.

Engelbrecht, J.A., Woodburn, E.T., 1975. The effect of froth height, aeration rate and gas precipitation onflotation. J. S. Afr. Inst. Min. Metall. 76, 125–132.

Espinosa-Gomez, R., Yianatos, J.B., Finch, J.A., Johnson, N.W., 1988. Carrying capacity limitations inŽ .flotation columns. In: Sastry, K.V.S. Ed. , Column Flotation ’88. SME-AIME, pp. 143–148.

Feteris, S.M., Frew, J.A., Jowett, A., 1987. Modelling the effect of froth depth in flotation. Int. J. Miner.Process. 20, 121–135.

Finch, J.A., Dobby, G.S., 1990. Column Flotation, Pergamon.Finch, J.A., Yianatos, J., Dobby, G., 1989. Column froths. Miner. Proc. Extr. Metall. Rev. 5, 281–305.Hemmings, C.E., 1981. On the significance of flotation froth liquid lamellae thickness. Trans. Inst. Min.

Metall., Sect. C 90, C96–102.Johansson, G., Pugh, R.J., 1992. The influence of particle size and hydrophobicity on the stability of

mineralized froths. Int. J. Miner. Process. 34, 1–21.Klassen, V.I., Mokrousov, V.A., 1963. An Introduction to the Theory of Flotation. Butterworth, London.Kosick, G.A., Freberg, M., Kuehn, L.A., 1988. Column flotation of galena at the polaris concentrator. CIM

Ž .Bull. 81 920 , 54–60.Laplante, A.R., Toguri, J.M., Smith, H.W., 1983a. The effect of air flow rate on the kinetics of flotation: Part

1. The transfer of material from the slurry to the froth. Int. J. Miner. Process. 11, 203–219.Laplante, A.R., Toguri, J.M., Smith, H.W., 1983b. The effect of air flow rate on the kinetics of flotation: Part

2. The transfer of material from the froth over the cell lip. Int. J. Miner. Process. 11, 221–234.Livshits, A.K., Dudenkov, S.V., 1965. Some factors in flotation froth stability. In: Proc. of VIIth Int. Proc.,

Longr. Gordon & Breach, New York, NY, pp. 597–621.Ž .Lovell, V.M., 1976. Froth characteristics in phosphate flotation. In: Fuersteau, M.C. Ed. , Flotation A. M.

Gaudin Memorial, Vol. I. AIME, New York, NY, pp. 597–621.Luttrell, G.H., Weber, A.T., Adel, G.T., Yoon, R.-H., 1988. Microbubble flotation of fine coal. In: Sastry,

Ž .K.V.S. Ed. , Column Flotation ’88. SME, Littleton, CO, pp. 205–211.Mankosa, M.J., 1990. Scale-up of microbubble column flotation. PhD Dissertation, Department of Mining and

Minerals Engineering, Virginia Polytechnic Institute and state University, Blacksburg, VA.Mehtrotra, S.P., Kapur, P.C., 1974. The effects of aeration rate, particle size and pulp density on the flotation

rate distributions. Powder Technol. 9, 213–219.Moudgil, B.M., Gupta, D., 1989. Flotation of coarse phosphate particles. In: Chander, S., Klimpel, R.R.

Ž .Eds. , Advances in Coal and Mineral Processing Using Flotation, pp. 164–168.Moys, M.H., 1984. Residence time distribution and mass transport in the froth phase of the flotation process.

Int. J. Miner. Process. 13, 117–142.Ž .Moys, M.H., 1989. Mass transport in column flotation. In: Laskowski, J.S. Ed. , Frothing in Flotation.

Gordon & Breach, pp. 203–228.


Parekh, B.K., Groppo, J.G., Bland, A.E., 1986. Optimization studies of column flotation for fine coal cleaning.In: Proceedings of the 3rd Pittsburgh Coal Conference, Pittsburgh, PA, Sept. 8–12.

Szatkowski, M., 1987. Factors influencing behavior of flotation froth. Trans. Instn. Min. Metall., Sect. C 96,C115–122.

Szatkowski, M., Freyburger, W.L., 1985. Kinetics of flotation with fine bubbles. Trans. Inst. Min. Metall.,Sect. C 94, C61–70.

Tomlinson, H.S., Fleming, M.G., 1965. Flotation rate studies. Proceedings, 6th Inter. Min. Proc. Congress,Cannes. Pergamon, Oxford, pp. 563–579.

Warren, L.J., 1985. Determination of the contribution of true flotation and entrainment in batch flotation tests.Inter. J. Miner. Process. 14, 33–44.

Xu, M., Finch, J.A., 1989. Effect of sparger surface area on bubble diameter in flotation columns. Can. Metall.Ž .Q. 28 1 , 1–6.

Yianatos, J.B., Finch, J.A., Laplante, A.R., 1987. Cleaning action in column flotation froths. Trans. Instn. Min.Metall., Sect C 96, C199–205.

Yianatos, J.B., Finch, J.A., Laplante, A.R., 1988. Selectivity in column flotation froths. Int. J. Miner. Process.23, 279–292.

Yianotos, J.B., Laplante, A.R., Finch, J.A., 1985. Estimation of local hold-up in the bubbling and froth zonesŽ .of a gas liquid column. Chem. Eng. Sci. 40 10 , 1965–1968.

Ynchausti, R.A., McKay, J.D., Foot, D.G., 1988. Column flotation parameters — their effects. In: Sastry,Ž .K.V.S. Ed. , Column Flotation ’88. SME-AIME, pp. 157–172.

Yoon, R.H., Luttrell, G.H., Adel, G.T., Mankosa, M.J., 1989. Recent advances in fine coal flotation. In:Ž .Chander, S., Klimpel, R. Eds. , Advances in Coal and Mineral Processing Using Flotation. SME,

Littleton, CO, pp. 211–218.

a parametric study of froth stability and its effect

Documents