Column Flotation 44-63

8/13/2019 Column Flotation 44-63

1/20

FROTH FLOTATION : RECENT TRENDScr?lIME, JAMSHEDPUR, 1998; pp_ 44-63Column Flotation Theory and

PracticeS.R.S. SASTR I

Regional Research Laboratory, Bhubaneswar - 751 013

ABSTRACTColumn flotation which employs a counter-current flow of slurryand air bubbles has proved to be a better alternative to theconventional mechanical cells for separation of minerals.Because of the distinct advantages of column flotation overmechanical cells, it is gaining wider acceptance in industry.Realizing the potential of column flotation, RRL Bhubaneswaralso took initiative in the early sixties to develop column flotationtechnology for Indian ores. In the early stages, operation anddesign of columns were mainly based on experience. But withthe increasing commercial installations, systematic investigationshave been carried out by several workers. In this paper anattempt is made to review the development of column flotation forconcentration of low grade ores. The basic principles andapplications of column flotation have been described. The salientresults obtained at RRL, Bhubaneswar have been highlighted.Key Words : Column flotation, Theory & practice, Applications.

INTRODU TIONSeparation of valuable minerals from gangue is far from ideal inconventional mechanical cells. Column flotation, invented in the earlysixties, proved to be a better alternative to the conventional cells. Themain advantages of column flotation are : i) improved recovery,ii) higher grade, iii) lower capital and operating costs, iv) less wear andtear due to absence of moving parts and v) requirement of less floorspace. Columns of varying design are in use all over the world 1 1 .Fig.1 shows a schematic diagram of flotation column. From anoperational point, two main zones can be identified : i) collection zone,where feed entering 1-2 m below the top of the column flows downcounter current to bubbles rising from a gas sparger near the bottom

44


2/20

S.R.S. SASTRIof the column and ii) a cleaning zone, where the froth rising from thecollection zone is washed of the entrained gangue by counter currentwash water introduced at the top of the column. Finally, the washedfroth overflows from the launder and is collected as the product whilethe tailings are discharged from the bottom. Commercial flotationcolumns can have either square or round cross-section and aregenerally upto 15 meters tall. Table 1 gives some of the early activitiesrelated to column flotation. Significant differences exist between thedesign and operating philosophies of mechanical cells and the flotationcolumns which lead to the difference in their performance. These aresummarized in Table 2.In the early stages, operation and design of the columns were mainlybased on experience. With the increase in commercial installationsseveral investigators particularly Dobby, Finch and co-workers 1 4 - 1 7 ] andYoon and co-workers 1 8 - 2 0 carried out systematic studies on the designand operational aspects. It is now possible to carry out theseoperations on a more scientific basis.In this paper an attempt is made to summarize the present status.

THEORETICAL BACKGROUNDSince there are no moving parts in the column the main operatingvariables are the flow rates : feed air, wash water, product and tailings.Column dimensions, bubble size, air holdup and the reagent dosagesare the other important variables. To normalize the effect of flow ratesin different sizes of columns, the superficial velocities, defined as thevolumetric flow rates per unit area of cross section, are used. Thenormal ranges of these are given in Fig.1. Since the effect of reagentdosages are similar to those in conventional cells, they are notdiscussed in the paper. Another variable which is generally mentionedwith reference to column operation is the bias rate which is thedifference between the tailings rate and the feed rate. If the value ispositive it is called positive bias and if it is negative it is called negativebias. Alternatively displacement wash ratio defined as the ratio ofquantity of wash water to the quantity of water reporting to theconcentrate.As mentioned earlier, the column can be divided into two distinctivelydifferent zones : the collection zone and the recovery zone or the frothzone. Both of these need separate treatment.

45


3/20

tails

frothzone, Hi

collect ionWM, He

S.R.S. SASTRIw ash w ate rJw0.05 - 0.3 cre/s

Fig. 1 : Schematic diagram of flotation columnCOLLECTION ZONEIn the collection zone the main factors affecting the recovery and gradeare the bubble size and air hold up besides the flow rates.Air Holding and Bubble SizeThe volume fraction of liquid displaced by air is known as the airholdup, E . The bubble size influences the hold up and the bubblesurface available for carrying the values.Except for clay type particles where viscosity effects dominate, in othercases slurry density and viscosity often have approximately equal andopposite effects on bubble rise velocity and therefore on holdup. It hasbeen shown 2 1 ) that bubble loading may result in significant increase inholdup, the increase being less significant for finer bubbles.

46


4/20

S.R.S. SASTRITable 1 : Early developments in column flotation

Year ctivity1962 nvention of flotation column.1963-67 est work at iron ore company of Canada and

Opemisaka Copper Mines (Quebec) on 0.45 msquare columnt2 t.

1966 irst publication giving results of tests at OpemisakaCopper Mines 3 1 .First publication from Regional Research Laboratory,Bhubaneswar 4 .

1971 aper in Russian on byproduct molybdenum recovery 2 1 .Publication of R&D paper on graphiteN.Publication of R&D paper on concentration ofmolybdenum oret6 1 .

1975 arallel testing of 0.45 and 0.9 m column at iron orecompany of Canada 2 1 .

1980 irst commercial column at Mines Gaspe 0.51 msquare column for Mo cleaning 7 1 .

1984 irst home made column 0.9 m dia at Gibraltar Minesfor Copper Cleaning 8 1 .

1986 cale up at Mount lsa Mines for Pb-Zn flotation 9 1 .Commissioning of 3 stage circuit at Gibralta Mines forbulk Cu/Mo cleaning 1 1 .

1987 ymposium on column flotation cell Trail B.C. Canada. 1 1 1 .1988 olumn flotation' 88 SME/AIME,[ 1 2 [ .1990 ook 'column flotation of JA Finch and G.S. Dobby 1 3 1 .

Effect of Gas and Liquid Rates on Hold upAccording to Shah et alt

2 2 ]the relationship between hold up and airsuperficial velocity defines the flow regime. The general trend is shown

in Fig. 2.It can be seen that the air hold up increase approximately linearly in thebeginning and then deviates above a certain J9 . The linear section ischaracterized by uniform distribution of bubbles, nearly uniform in size,and is known as bubbly flow regime. This is the region of interest in

47


5/20

Mechanical cells lotation columns1. Similar to ideal mixers perate under conditions

of plug flow with varyingdegrees of axial dispersion

2. Air bubbles are formed by a ir bubbles ore formedrotating impeller y passing compressed

air/air slurry mixturethrough a bubblegenerator

3. Relative velocity between air elative velocity betweenbubbles and mineral particles ir bubbles and mineralin negligible except near the articles is high throughimpeller. Hence chances of ut the length of thecollision ore reduced olumn since they

move counter current4. At any given time only a small otal length of collection

fraction of the mineral particles one is available foris in the viscinity of air bubbles ollision and attachment.'created. Thus effective hus total residence timeresidence time of particles is s effectively utilizedsmall compared to the totaltime of presence in the cell

5. The highly turbulent conditions he quiescent operationpromote i) detachment of once esults in i) reducedattached particles and ossibility of detachmentii) contamination of froth by nd ii) reduction inentrainment of non floatable ntrainment of gangueparticles inerals in the froth

6. Relatively large size bubbles elatively smaller bubblesnot favorable in flotation of ive higher surface andfine particles igher residence times.

7. Addition of wash waterfurther improves the gradeof the product i) by pushingdown the process watercontaining gangue mineralfrom going with the productand ii) washing down thecoarse gangue particlescarried over to the froth byentrainment.

S.R.S. SASTRITable 2 : Comparison of operating mechanisms of conventional cells and

flotation columns

48


6/20

S.R.S. SASTRIcolumn flotation. Above this air rate, hold up is characterized by largebubbles, rising rapidly displacing water and small bubbles downward.This region is known as the churn turbulent region and the operationwill be unsteady in this region.For a given flow rate of air, increase in counter current flow of waterincreases hold up as a result of relative decrease in bubble risevelocity. However, increase of J, will decrease the maximum Jg forbubbly flow regime.

Gas hold-up

Superficial gas velocity

Fig. 2 : General relationship between gas rate and hold up

Addition of frotherAddition of frother up to a certain level has a pronounced effect ofreducing bubble size, resulting in reduced bubble rise velocity andconsequent increase in hold up.Estimation of Bubble SizeIt is difficult to measure the bubble size in operating columns. Basedon laboratory studies, it is possible to estimate the bubble size using

49


7/20

S.R.S. SASTRIdrift flux analysis[ 2 3 i. Accordingly to this method the data required arethe hold up (and liquid and air rates J and Jg .The steps involved areI) assumption of a db valueii) calculation of slip velocity (relative velocity) us from

us = (Jg /e) + J/(1-E) 1)iii) calculation of Reynolds number ResRes = db u p (1-E)/p. 2)iv) calculation of the terminal velocity u

= u 41-02 3)v) and calculation of bubble size using the equation

db = [1 81..Lu t (1+0.15 Res a6 8 7 )/g pr 4)and iterate till db calculated is equal to db assumed.Limiting ConditionsFrom equations (1) and (4) it follows that, for a given J there is arestriction on the permissible Jg , db combination. Under normaloperating conditions in flotation columns, over a db range of 0.6 to 1.2mm, the maximum superficial bubble surface rate S defined as bubblesurface rate per unit cross section

Sb = 6Jg /db 5)Sb is found to be independent of bubble size [ 1 4 1 . It means that themaximum permissible gas rate decreases with decreasing bubble size.The implication of this is that decreasing bubble size may not improvesolids removal rate. Instead column may be operated at higher Jg rateskeeping the bubble size near the upper limit to improve solids removalrate.Collecting of ParticlesThe fractional recovery of a mineral particle in the collection zone isgiven by

R = 1-exp. (-kb tp ) 6)for a first order rate process under plug flow conditions which isnormally the case in laboratory columnsThe first order rate constant ice is given by

kc = 1.5 Jg EK /db = 1.5JEc E /db 7)50


8/20

S.R.S. SASTRIassuming particle detachment to be negligible due to absence ofmechanical agitation. Studies have shown that

Ec a do '/db 8)where m varies from 1-2 and n varies from 2.5-3 and EA decreases withincreasing particle size and increases with increasing particle density.For column with internal sparger Jg and db are dependent, Equations(7) and (8) show the increasing Jg decreases k, by increasing db andreducing Eo .An advantage of external spargers over internal spargers is that Jg canbe increased independent of db in the former to increase IceMixing in Collection ZoneLaboratory flotation columns typically operate under plug flowconditions while plant columns operate under conditions intermediateto plug flow and perfectly mixed flow. For these columns plug flowdispersion model is shown to provide a good working basis.The axial dispersion is commonly quantified by a dimension lessnumber known as Peclet number. Mankosa et al'1 9 1 proposed theequation

Pe = 0.7 (H/D) 3 (ul/Jg )0 5 9)for estimating the degree of dispersion in flotation columns. For plugflow conditions Pe is infinity and for fully mixed conditions Pe is zero.According to Eq. (9) dispersion in flotation columns can be decreasedby increasing H/D, interstitial liquid velocity u, or decreasing JgOver the bubble size range relevant to column flotation, decrease ofbubble size is reported to increase dispersion.Particle Residence TimeMean particle residence time reduces with increased particle size andincreases with interstitial liquid velocity. It can be calculated from thefollowing equations by iteration

to = t, [J, /(1-E)]/ [us p + J /(1-e)] (10)where us p = gdp 2 (pp - ps1 ) 1-0)2-7 /181 .1. 1+0.15 Re s 8 7 ) (11)

Rep = p us p p(1-0)4t (12)and i H (1-E) / J (13)

51


9/20

S.R.S. SASTRILocation of Feed PointLocation of feed point is an important factor since it has a bearing oneffectiveness of collection zone. Having a feed point in the lowerportion is likely to result in short circuiting of solids to the tailings andalso gives less residence time to the solid particles for effectivecollection. On the other hand having it too close to the interface is likelyto disturb it. A compromise is to have the feed point about 2 meter fromthe top in commercial columns.Effect of Mixing on RecoveryRecovery in a plug flow column with mixing can be calculated from theequation

R=100 [1-4a exp (Pe/2)/(1+a2 )exp (a Pe/2)-(1-a2 )exp (-a Pe/2)](14)

where a = (1+4ktp / Pe)0 5 15)and t is calculated from Equations (10-13)

As mentioned earlier, the Peclet number decreases as the columndiameter increases and recovery would decrease unless tp isincreased. Increase of t means increase of column height which willresult in change of H/D and consequently changed dispersion. Underthese conditions t required for the same recovery will change. Due tointerdependence of the variables, iterative process is needed tocalculate the H or t required for a given recovery in commercialcolumns. Mankosa 1 9 1 et al have illustrated this effect by plotting theratio of mean residence time required in the large column to that in thelaboratory column as a function of 'column diameter for different flowconditions.For a given residence time, increase of H/D ratio will also result inreduced volumetric flow rate ( to maintain similar J) which means inoverall reduction of collection of solids. Increase of gas rate generallyresults in improved recovery but lower grade of the product.Bubble GeneratorsBubble generators are termed as the hearts of flotation column. Thesecan be divided into two groups : i) internal and ii) external. In the earlystages of development only internal spargers are used but at presenttheir use is limited to laboratory and pilot test units. The metallurgicalperformance of these two types of spargers are reported to besimilar 2 4 ]. The advantages and disadvantage of these project are givenin Table 3.

52


10/20

S.R.S. SASTRITable 3 : Advantages and disadvantages of different types of spargers

Advantages isadvantagesExternal spargersi. Less chances of plugging . Proprietary items and relatively

costlyii. On line maintenance i. Operation relatively complicatediii. Control over bubble size ii. Need other accessories like

pump or high pressurecompressor

iv. Long lifeInternal spargers1. Relatively cheap . On line maintenance not possible

leading to production loss2. Can be fabricated locally . Need to change once in 3

months3. Operation is easy . Control of bubble size not

possible4. Require low pressure air

Internal spargers of different shapes were in use. These includeperforated pipes covered with perforated rubber or filter cloth, disk filterelements or inverted cone type covered with filter cloth, and porousmetal spargers.In case of the above type of spargers the bubble size is found to berelated to the gas rate and relative areas of sparger and column Rsaccording to the relation

db C (Rs Jg ) 16)where C depends on frother concentration for a given set up.Typical examples of external sparger are USBM, Cominco, Minnovex

and Microcel spargers (Fig. 3).FROTH ZONEThe holdup of air in column froths is around 80%. Increase of J9beyond a limit tends to increase hold up in collection zone and reducethe hold up in the froth zone with consequent loss of interface which

53


11/20

EAD ADJUSTER

air

COLUMN WA LL

MOVA BLE HEA D air je t 200 - 400 m/s)

bubble - s lur ry dispers ion

S.R.S. SASTR I

air (a)

w ter

b)

gas B

C)

Fig. 3 : External spargers(a) USBM/Cominco, b) Minnovex c) Microcel

54


12/20

S.R.S. SASTRIis not desirable for smooth column operation. In order to reduce theeffect of entry of more feed water into the froth zone with increasing Jgmore wash water needs to be used.Plant experience indicates that froth depth has no significant effect onmetallurgical results.Wash WaterThe use of wash water distinguishes the column froth from theconventional cell froth. The purpose of addition of wash water is i) toprovide water necessary for the overflow of the collected solids into thelaunder and ii) to suppress the water coming from the feed from goingalong with the product in order to prevent the carry over of gangueminerals by entertainment.Froth DropbackDropback from froth is important to calculate the overall recovery.Measurement of this is very difficult and the limited data availableindicate that the froth drop back varies widely from about 20 to80 .The effect of different variables on the processes occurring in theflotation column were dealt with so far.To summarize, the column performance is greatly influenced by anumber of variables as shown in Table 4.Now, the practical aspects of testing and design will be dealt with.

Table 4 : Influence of operating variables on column operationVariable ffected propertyAir rate ubble size, holdup, kinetics, carrying

capacity, dispersion, product gradeand recovery

Feed rate ubble size, dispersionBubble size oldup, kinetics, carrying capacity, dispersionWash water ntrainmentParticle size esidence time, kineticsViscosity oldup, kineticsH/D ratio ispersion

55


13/20

S.R.S. SASTRITestingTesting has multiple objectivesi Demonstration of feasibility or amenabilityii) Determining the range of operating variables and parameter

estimationiii) Collection of engineering data for scale upAmenability TestsAmenability tests are carried out to establish grade-recovery curves forcomparison against standard laboratory test results or existing plantperformance. The variables generally studied are the residence time,

solids in feed and gas rate. Feed rates may be controlled usingperistaltic pumps. Interface is maintained manually by adjusting tailingsrate. A period of 3 residence times is normally allowed before collectingsamples.If the column gives superior results compared to those from thelaboratory mechanical cell the inference is that column is better suitedsince laboratory mechanical cells often produce better resultscompared to the plant size cells.Parameter EstimationRate constantGenerally the over all rate constant k (obtained in presence of froth) ismeasured instead of the collection zone rate constant Kc (assumingfroth zone recovery to be 100 ). For measuring this, the levels of othervariables like air rate, reagent dosages which effect the rate constantshould be fixed. Care should be taken to see that the air rate used issufficient to keep it away from the fully loaded condition.The overall rate constant k can be determined by varying the tailingsrate in a long column (approx. 10 M long) or recycling the tailings in ashort column. A disadvantage of the latter method is the possibility ofsurface modification due to repeated handling.K the rate constant is estimated from the slope of In (100-R) vs.residence time plot.The collection zone rate constant lc, can be estimated directly byoperating the column at high bias rate to eliminating the froth zone andmaintaining a low level of recovery by entertainment.

56


14/20

S.R.S. SASTRICarrying capacityCarrying capacity is defined as the concentrate removal rate in termsof mass of solids overflowing per unit time per unit column crosssectional area. It is shown that this can beestimated from theequation2 5 ].

Cm a . = 0.049 dp p s 17)Since this gives the upper limit where bubbles are fully loaded withmineral particles, the normal operating capacity should be below thisand a reasonable estimate for the operating level is given byI 5 1.

C = 0.03 dp p s 18)The carrying capacity is experimentally determined by operating thecolumn at a given retention time varying the feed solids rate (throughfeed percent solids) till the maximum in concentrate solids rate isachieved.Hold up in the collection zoneA practical way of measuring air hold up is by using pressuretransducers located at two different heights in the collection zone ( Fig.4). The hold up can be calculated from the equation

e = AP /p s i gAL 19)To reduce the derivations in the estimated hold up, AP is measuredover a section near the bottom of the column (above the air sparger)where the bubbles are expected to be only lightly loaded and thetailings density is used as an approximation to the slurry density withinthe region.

Liquidlevel

IIAL AP

a

IAirFig. 4 : Measurement of gas hold-up

57


15/20

S.R.S. SASTR IInterface between froth and collection zoneThe location of froth slurry interface is normally carried out by the useof differential pressure measurement between two or more locationsabove the feed level.Pilot plant testingPilot plants of 0.2 -1.0 diameter are constructed as an intermediatestage. The main objectives of the pilot plant are i) checking the resultsof amenability tests and the preliminary scale up data, ii) testing andevaluation of practical aspects and process control instrumentation andiii) operator training.CONTROLFor stabilized operation and optimum performance controlinstrumentation is required. Among the different types of stabilizingcontrols used, the simplest one is to control the interface level bymanipulation of tailings rate. In this system, wash water addition ismanual with and no control over bias. A deep froth is generallymaintained to dampen the effects of gas and bias rates.In an alternate method of control, wash water is manipulated to controlthe level and tailings rate to control the bias.Both methods are reported to give similar metallurgical performance,but the former is relatively simpler.Current information available leads inadequate understanding of theeffects of air rate, hold up, bias, wash water rate and froth depth onmetallurgical performance and it is difficult to suggest a general systemof control instrumentationSCALE-UPScale up of flotation columns is generally based on kinetic modelsusing axial dispersion theory. The models proposed by Finch andDobby and co-workers[ 1 3 2 6 1 and Yoon and co-workers 1 8 - 2 0 1 areprominent among these. Laboratory columns operate under plug flowconditions, while the plant columns operate under conditionsintermediate to plug flow and perfectly mixed conditions.The over all recovery, R is given by

R = 100 IR,R1- Rc + R 20)where Fic and R, refer to recoveries in collection zone and froth zone

respectively.58


16/20

S.R.S. SASTRIThe froth zone mechanisms are not fully understood, but indicationsare that the froth zone recovery may vary between 20-80 with anaverage of 50 .For partially mixed flow condition, is given by Eq. (14).Column DiameterThe diameter of the column is estimated from the data on superficialvelocity of feed and the carrying capacity.Height of the Collection ZoneThe height of the collection zone for a given recovery can be estimatedfrom Eqs. (9-15). The dispersion equation (Eq. 10) indicates thatincrease of H/D decreases dispersion. For the same retention time, itis possible to increase H by reducing D. But varying column geometrycauses other changes in other variables like particle velocity, bias andgas flow rates, besides the capacity which depends on the crosssectional area. It can be seen from the relationship.

Q = J V /H 21)where Q is volumetric air rate, increasing H at constant column volumeresults in lower air rate. This means increased bubble loading withincreasing H/D rites. Beyond a point further increase in H/D may notbe beneficial since the bubbles are already fully loaded. There is apractical limit to this ratio.It is interesting to note that the performance of commercial plant isfound to be superior to the one obtained by simulation using data fromthe laboratory and pilot plant test 9 . 2 7 1.APPLICATIONSSince the flotation column was tested in the sixties on iron ore andmolybdenum, the first commercial application was replacement of anumber of stages of cleaning in molybdenum circuit. Originally most ofthe applications were in replacing a number of stages of cleaning inmolybdenum, lead-zinc circuits by column. Column flotationapplications have increased covering roughing and scavengingalso 2 8 2 9 1. Even all column flotation installations are reported proving theversatility of columns.

59


17/20

S.R.S. SASTRI

a) b)

c) d)Fig. 5 : Some industrial Column Flotation projects of RRL Bhubaneswar a)1 m dia Column West Bokaro coaking coal washery Tata Steelb) Molybdenite recovery at Rakha Copper concentrator c) 1 75 m dia Columnsn Zn cleaning circuit Dariba concentrator, Hindusthan Zinc Ltd. andd) Molybdenite recovery at Uranium Corporation of India Ltd., Jaduguda

60


18/20

S.R.S. SASTRICOLUMN FLOTATION : REGIONAL RESEARCH LABORATORYBHUBANESWARRealizing the potential of column flotation, the laboratory took initiativein the early sixties to adopt it to the Indian ores and to develop designcapabilities. An inverted cone covered with filter cloth was used as theair sparger. A wide variety of materials graphite, coking coal, limestone,sillimanite, molybdenite, Zn-rougher concentrate were tested usingcolumns ranging from 0.056 to 0.22 meters diameter. Based on theamenability tests carried out at the laboratory, a one meter diametercolumn to process 3-4 tones/hr of tailings of present flotation plant wasset up at West Bokaro Coking Coal Washery of Tata Steel. Thelaboratory is also associated with the first indigenous effort in scalingup and operation of the columns in the zinc cleaning circuit at Daribaplant of Hindustan Zinc Ltd. Preliminary results indicate that theperformance may be even better than the targets set in design. Thelaboratory also taken to some of the earliest publications on columnflotation. Fig. 5 shows some of the industrial column flotation projectswith which the laboratory is associated.NomenclatureC Carrying capacity normal operatingC. Maximum carrying capacityD Diameter of columndb Diameter of bubbleEA Attachment efficiencyE Collision efficiencyE Collection efficiencyg cceleration due to gravityH eight of collection zoneJ8 Superficial velocity of bias

Superficial velocity of feedJ uperficial velocity of airJi uperficial velocity of liquidJS I Superficial velocity of slurryuperficial velocity of tailingsJw Superficial velocity of wash,

waterk verall rate constantk c Collection zone rate constantAL Distance between two points for

measuring differential pressureAP Pressure differential between

the two points

Peclet numberOverall recoveryCollection zone recoveryFroth zone recoveryReynolds number of bubbles inswarmReynolds number of particleMean residence timeLiquid residence timeParticle residence timeLiquid interstitial velocityParticle interstitial velocitySlip velocity between air bubbleand waterTerminal velocity of air bubble

ek LettersAir holdupViscosity of waterViscosity of slurryDensity of liquidDensity of solidsDensity of slurryVolume fraction of solids in slurry

PeRRcR,RebRep

tP

us

u,GreE

PP SPsi(

61


19/20

S.R.S. SASTR IREFERENCES1. Reddy, P.S.R., Kumar, S.G., Bhattacharya, K.K., Sastri, S.R.S. and

Narasimhan, K.S., 1988. Int. J. Miner. Process, 24, p. 161.2. Wheeler. D.A. Column flotation - The original column paper presented at

the 87th Annual General Meeting of CIM-Van Coevur, 24th April 1985.3. Wheeler, D.A., 1966. E&MJ 167 (11), p. 98.4. Choudhury, G.S. Lakshmanan, C.M., Narasimhan, K.S. and Rao, S.B.,

1966. The Explorer, p. 1.5. Narasimhan, K.S, Rao, S.B. and Choudhury, G.S., 1972. E&MJ 173(5),p. 84.6. Mathieu, G.I. , 1972, CIM Bulletin, 65, (271), 1972, p. 41.7. Coffin, V.I., 1982. In : Procd. XIV International Mineral ProcessingCongress, Toronto, p. IV-21.8. Mauro, F.L. and Grundy, M.R., 1984. In : Procd. CIM District Meeting,

Kamloops, B.C.9. Espinosa-Gomez, R., Johnson, N.W. and Finch, F.A., 1989, Minerals

Engg. 2(3), p. 369.10. Redfearn, M.A. and Egan, J.R., 1989. In : Procd. International Symposiumon Processing of Complex Ores, G.S. Dobby and S.R. Rao (Eds.),

Pergamon Press NV, p. 303.11. Feasby, D.G. (Ed.), 1987, In : Procd. Column Flotation Cell Symposium,

B.C. Canada, CANMET Ottowa.12. Sastry, K.V.S. (Ed.), 1988, In : Procd. Column Flotation 88, SME/AIME,

12th Annual Meeting, Phonix, Arizona.13. Finch, J.A. and Dobby, G.S., 1990, In : Column Flotation, Pergamon Press

Oxford.14. Finch, J.A. and Dobby, 0.3., 1991, Int. J. Min. Process, 33, p. 343.15. Dobby, G.S. and Finch, J.A., 1991, Minerals Engg., 4 (7-11), p. 911.16. Finch, J.A. Uribe-Salas, A and Xu; M., 1995, In : Flotation Science and

Engineering, K.A. Matis (Ed.), Marcel Dekker New York, p. 291.17. Finch, J.A., 1995, Minerals Engg., 8(6), p. 587.18. Mankosa, M.J., Adel, G.J., Luttrel, G.H. and Yoon, R.H., 1990. Mineral

and Metallurgical Processing, Rajamani, R.K. and Herbst, J.A., (Eds.),Salt Lake City, USA, SME 35.

19. Mankosa, M.J., Luttrel. G.H. Adel, G.J., and Yoon, R.H., 1992, Int. J.Miner. Process., 35, p. 51.

62


20/20

S.R.S. SASTRI20. Yoon, R.H, Mankosa, M.J. and Luttrel, G.H., 1993, In : Procd. XVIII

International Mineral Processing Congress, Aus/MM, Sydney, Australia ,p. 785.

21. Yianatos, J.B., Finch, J.A. Dobby, G.S. and Xu, M., 1998, J. Coll. Inter.Sci. 26(1), p. 37.22. Shah, Y.T. Kelkar, B.G., Godbole, S.P. and Deckwer, W.D., 1982, AIchEJ

28 (3), p. 353.23. Banisi and Finch, J.A., 1994, Minerals Engg., 7(12), p. 1555.24. HuIs, B.J. Lachance, C.D. and Dobby, G.S., 1990, Minerals Engg. 4(1),

p. 37.25. Sastri, S.R.S., 1986, Minerals Engg., 9(4), p. 465.26. Falustu, M., and Dobby G.S., 1989, In : Procd. Complex Ores, Dobby,

G.S. and Rao, S.R., (Eds.) C.I.M Halifex, Canada, p. 335.27. Hall, S.T. and Averiss, S.B., 1988, In : Procd. Fine Particles, Plumpton, A.(Ed.), Can Inst. Min. Metall, p. 181.28. Newell, A.J., Cantrell, R.A. and Dunlop, G.A., 1992, In : Procd. Extractive

Metallurgy of Gold and Base Metals, Kalgorlie, Inst. Min. Met., p. 253.29. Brewis,T., 1991, Mining Magazine, p. 383.

63

Documents

Column Flotation 44-63