V man , A. R. , and Iwasaki . I . , 1968 . "MineraJogical andBeneficiation Studies of the Copper-Nickel Bearing DuluthGabbro," 29th Annual Mining Sym~um. Center forOxltinuation Study. Uni\'ersity of M~n_~, Minneapolis.MN.

Watton. I.. 1980. "Olivine and Dunite." Industrial Min.erals. No. 159, p. 57.

Weiblen, P.W., and Morey, G.B., 1976, "TextUral amCompositional Characteristics of Sulfide Ora &om thecBasal Qmtact Zone of thec South Kawishiwi intrUsion,Duluth Complex, nonheastem ~~~~, 57th AnnualMining Symposium, Center for Qmtinuation Study, Univer.sity of Minnesota. Minneap>lis, MN.

lJobik, A.B.. 1981, "Mica," Mining Engineering, Vol. 55.No, 5, p. 580.

8.M. MO\Kfgi1 and P. Sc.naSW1daran

~ - Adsorption of a nonionic pol)dcryiamid8

(PAM), an anionic poi)laC1)'lamid8 with sulfonate functJ"on-al groups (PAMS), and a cationic pol)dcryiamid8 withaminefunct ion4l groups (PAMD), on hem4tite ~ ~-ligated as a.function of different solution conditiO1LS.Adsorption results indicate that the major mechanisms ofadsorption of pol)dcryiamides on hem4tite if throughhydrogen bonding, with el.ctrosta.lic forces ~ aseCondilry rol.. AU the pol)lmer.f exhibited irreversibility ofadsorption upon dilution. Also, entropic effects were deter-mined to be important in adsorpcion. Surface chemicalproperties of hem4tite and solution chemistry of the poly-acryiamides are considered to exPlain the polymer adsorp-tion behatior on hem4tite.

principl~ that goYem the effect of such interaCtions inmineral procesing, effect of solution conditions on theadIOtption of a nonionic polyacrylamide (PAMS), anan¥)nic polyacrylamide (P AMS - sulfonate fwxtional group).aOO a cationic polyacrylamide (PAMD-amine functionalgroup) on hematite was investigated.

Badtgrotmd

Adsorption of Polyacrylamides on Hematite

AdMJrption of a PJIymer on a mineral surface is attnootedto electrostatic charge attraction, hydrogen bonding,covalent bonding, or hydrophobic interactiom. It has beenreported that the principal mechanism for the ad8orption ofpolyacrylamide type PJlym~ on oxide minerals is throughhydrogen bonding (Linke and Booth, 1959). Hydrogen is~ between surface oxygen atoms and oxygen or~ of the }X)lymer. Such adlorption will be influencedby any charge in the PJlymer backbone, anionic or cationic(by hydrogen bonding). In this case, modWcatiom in thr:suspension's pH will ~t in: alterations in surface chargecharacteristics of hematite since H+ and OH- are thr:potential deterD1ining ions; and, changes in the ionic natureof the PJlyacrylamides due to hydrol}'Sil, especially underbasic pH conditiom. A brief discu:Bion of the abo\Oe twofactors is presented below.

httroduction

Beneficiation of mineral fines using floc flotation tech.nique depends on the adsorption of the polymeric fIoccuiantonly on the dcsired mineral component. One method ofachieving ~ polymer adaorption is by controlling theinteractions between the polymer and the surfactants ~as collectors. These interactions could not only modify theadIOrption behavior of various molecular specie on themineral particle. but can alto lead to precipitation, ~tingin higher reagent consumption (Somasundaran andMoudgil, 1982). Precipitation can aJao ad~ly affectseparation efficiency of the proces. At present, effectiveaeparation of clays from sylvite at Cominco in Canada isachieved commercially through controlled interactionbetween the polymeric and collector species. It would.Mwever. be nccrsary to undentand the mcch2nism of suchinteractions before this phenomenon could be successfullyapplied to beneficiate other complex ora. To study the

Surface Chemical Properties of Hematite

~ isoeJectric point (iep) of synthetic F~O5 (hematite)UIed in this study was detem1ined by electrophoreticmeasurements using a %eta meter to be pH 8.1. ~ iep andF (point of zero ~) valua for h)-drated iron oxide~pitates by slow and fast acid-b~ titration methodshave been reponed to be in the pH range of 8.S to 9.S(Parks and deBruyn. 1962; Onoda and deBruyn. 1966; andAtkinson. P(81eI". and Quirk. 1967). Diff~ in the iepvalue obtained in this study and th~ reponed elsewherecan be due to the materials used. or. it could be due totechniqua eJDpio)'t:d fordet ermining the iep and F val\D.A detailed discusion of decaod1emical piufJt;i1ics of theoxide solution interface is p~ted by Ahmed (1975).

8.M. Moudgll, member SME, Is an associate professor in theDepartment of Materials Science and Engineering at theUniversity of Florida, Gainesville, FL. P. Some.undaran,member SME, is a professor at the Henry Krumb School ofMines, Columbia University, New York. SME preprlnt 82.160,SME.AIME Annual Meeting, Dallas, TX, Feb. 1982. ManuscriptMey 1982. Discussion of this paper must be submitted, Induplicate, prior to Aug. 31, 1_.

I--T;;,-~... Vol. m Soci8Iy of Mir8ng Engi- 01 AlME

_'._L'".~-'~ -

investigation on the nature of association of cobalt indifferent mineral p~. particularly in olivine andplagiocl~. become of interest.

RefereD(X!S

Table 5-Distribution of Nickel, Cobalt, and Sulfurin Bulk Sulfide Flotation Tailing

~ CDMlt

% c. Duo.

Ioalf-

IS _.% 81

0..7

0.09

0.02'

0.10

0.10

0.OS5

4.7-

0.063

0.060

~t

1 Ut.

47.8

2!.'

lS.'4.22.14.2G.!

l~.G

'lq14Cl...011.-'yr.-81otlt.*p.tlt.118-.1..S.lfU..

c-.o.lt.

_4_1.

0.013

0.02'

0.015

0.025

0.020

0.030

0.300

0.020

0.01'

31.0

36.0U.SS.S2.06.S

7.S

100.0

0.06

0.066

0.007

0.06

O.tO

0.06

31.'-

0.23

0.22

U.57.'8.51.18.21.1

"..118.8

'.73.33.7

37.4

100.0

.E " f.. .ho ,-ical '_01'- of ..lfUe f1o&..- ... (~ 3]

Bjoenun, H., 1975, "Olivine: An Interesting IndustrialMineral," Industrial Minerals, No. 99, p. 48.Boucher, M.L., 1975, "Copper-Nickel Mineralization in aDrill Core from the Duluth Complex of Northern Min-ncsota," US Bureau of Mines RISO84.Chace, F.M., and Stone, J.G., 1969, "Iron Ore and F1ux,"in J.H. Straaburger (ed.), Blast Furnace - Theory andPractice, Gordon and Breach, New York, p. 118.

Collings, R.K., 1980, "Mineral Waste Resources of CanadaRepott No.6 - Mineral Wastes as Potential Fillers," CanMet Report SO-lSE.

Flger, G.W., Kirby, D.E., Rhoa&, S.C., and Stickney,W.A., 1974. "Synthesis of Rutile from Domestic Ilmenites,"US Bureau of Mines RI7985.

Henn,J.J., and Barclay. J.A., 1970, "A Review of ProposedProce&1es for Making Rutile Substitutes," US Bureau ofMines 1C8450.

Iwasaki, I., Reid, K.J., Lex. H.A.. and Smith, K.A., 1985,"The Effect of Autogenous and Ball Mill Grinding onSulfide F1otation," Tram. SME-AIME, Vol. 274, pp. 1184-llQnSwnmaf)'

Laboratory concentration tests were run to study the})(8ibility of recovering byproduct minerals from flotationtailings of copper-nickel bearing Duluth gabbro. A flow-sheet involving gravity concentration. followed by low-intensity magnetic separation. sulfide flotation. andelectrical concentration of the gravity concentrates for theseparation of magnetite. residual sulfides. ilmenite. andolivine was developed. It also included high-gradientmagnetic separation and biotite flotation of the gravitytailings for the separation of plagioclase and biotite. Theflowsheet was tested on six Duluth gabbro sampl~ taken atfive different locations of the Ininera1ized area. Theflowsheet was shown to be relatively insensitive to theregional characteristics of the ore. The ~ of a Humphreyspiral in preconcentrating the heavy minerals ahead of ashaking table showed considerable promise in lightening theload to the shaking table and also to the high-gradientmagnetic separator for the concentration of plagioclase.

Only the ilmenite concentrates met commercial specifica-tions. Magnetite. plagioclase. biotite. and olivine wereeither too low in grade. or marginal. Attempts were made tofurther upgrade these mineral fractions and the limit ofpurity of each byproduct was ascertained. However. by-products recovered in the above flowsheet were rather snall.altogether amounting to 15 to 45% by weight. It becom~ ofinterest to ascertain the effect of closing the circuits on therecovery and on the purity of each byproduct. .

r~

6-

(I.:

J

tr

Acknowledgment

This material is based on work supponed by the Office ofSurface Mining. Department of the Interior under GrantNo. G519502S. Any opinions. findings. and conclusions orrecommendations expressed in this publication are those ofthe authors and do not nec~rily reflect the views of theOffice of Surface Mining. Department of the Interior.

Iwasaki, I., Malicsi, A.S., Lipp, R.j., and Walker, j.S.,1982, ResouTcesand COfISemztion, Vol. 9, pp. 105-117.

james, R.j., 1980, "The Use of ilmenite in Blast Furnaces,"IndustnaL Minerals, No. 155, p. 47.

Koenig, C.W., 1980, Ohio State University, personalcommunication.

Koiliari, N.C., 1974, "Recent Developments in ProcessingIlmenite for Titaniwn," International Joumal of MineralProcessing, Vol. I, pp. 287 -305.

Lawver,j.E., Wecmler, I., Arvidson, B., and OberteufIer,j.A., 1976, "Production of Anonhosite Concentrate fromMinnesota Copper-Nickel Flotation Tailings by High-Gradient Magnetic Separation," World Mining ~ MetalsTechnology, proceedings of joint MMIj-AIME meeting,Denver, CO, Vol. 2, pp. 929-942.

Lawver, j.E., Wiegel, R.L., and Schulz, N.F., 1976,"Study of Economic Feasibility of Copper-Nickel Develop-ment in Northern Minnesota," S7th Annual MiningSymposium, Center for Continuation Study, University ofMim1esota, Minneapolis, MN.

Lynd, L.E., 1980, "Titanium Minerals," Mining Engineet-ing, Vol. S2, No.5, p. 587.

Minnesota Environmental Quality Board, 1979, The Mm-nesota Regional Copper-Nickel Study, five volumes.

N.L. Industries, New York, Metallurgical Laboratory,1980, private communication.

Rogers, W.F., 1965, Compo.st"tion and Properties of OilWeU DnUing Fluids, Gulf Publishing, Houston, TX, p. 265.

Skz1lings'Mining ReTiew, 1975, "U.S. Steel to Close DuluthCement Plant," Oct. 4, p. 8.

Smith, K.A., Riemer, S.C., and Iwasaki. I., 1982,"Carbochlorination of Aluminum from Non-BauxiteSourceS,"Joumal of Metals, Vol. M, No.9, pp. 59-62.

US Bureau of Mines, Mineral Commodity Summaries 1980.

LSociety Of Mining EngIneers of AJME Tr8n18c1ioM VOl. 27&-1821

Solution Chemistry of Polyacrylamides

H)drolysis of polyacrylamide type polymers containingamide ( -CONH!) groups bas been reponed to occur under~ acidic and basic pH conditions (Higuchi and Senju.1972: Moens and Smets. 1957; Pinner. 1955: and Smets andH~bain. 1959). Since hydrolysis of amide groups by acidshas been reponed to be significant only at temperatureshigher than 100°C (212°F). it can be a.wumed that underexperimental conditions of 25°C (77°F) polymer structuredid not change in the acidic pH range. Kinetics of hydrolysisof polyacrylamidcs under basic pH conditions are fastenough to result in conversion of a significant number of-CONH! groups to -COOH groups in the followingmanner (Kulicke and Klein. 1978):

RCONH! + HOH ~ RCOOH + NHs

Thus. ~ng on the pH that governs the di8OCiation of-COOH groups. there could be changa in the ionic natureof the polyacrylamide type polymers. For example. uponhydrolysis. a nonionic polyacrylamide (PAM) can behave asan anionic polymer. whereas there can be an increase in theanionic nature of a sulfonated polyacrylamide (P AMS).01anges in the ionic nature of a cationic polyacrylamide(P AMD) as a function of pH are more complex:. The overallcharge of PAMD is based on the hydrolysis of the -CONH!groups to -COOH groups and. hydrolysis of the aminegroup (from DMAPMA-the comonomer) attached to thepolymer backbone to neutral RNH! groups. At higher pHvalues. therefore. positive charge density of the cationicpolyacrylamide will be reduced.

where [PJ] is the intrinsic viscosity, M is the average molecularweight, and 'K' and 'a' are constants for a given polymer~t system (Huggins, 1958). For polyacrylamide solutionin water, values of 'K' and 'a' are given as 6.51 X 108 and0.8, Ig-~-.:ively (Polymer Handbook, 1974). These con-stants are reponed to be valid for polymers of up to 500,000molecular weight. Values of OK' and 'a' for higher molecularweight polyacrylamides are not available. Therefore, theabove values were used to estimate average molecularweights in the range of one million or more. The Mark-Houwink relation was used to estimate molecular weights ofpolyacrylamide! with sulfonate and amine functionalgroups also. becaUR no other relationships for polyelectro-lytcs of this type are reponed. Since the amount ofaxnonomer in the polyetectrolytes is small (5 mol %). itmight be expected that the [PJ] versus M relationship will begoverned by the behavior of the backbone (polyacrylamide).This will be even more true if the contribution of the ionicfunctional groups towards the exten.ion of the polymermolecule was suppressed. This was achieved by measuringthe intrinsic viscosity of the ionic poiymen under high ionicstrength conditions. VISCosity average molecular weights ofthe polymers were detennined to be 2.5 million for nonionicpolyacrylamide (PAM), 2. 1 million for anionic polyacryla-mide with sulfonate functional group (PAMS). and 1.9million for cationic polyacrylamide with amine functionalgroup (P AMD).

Inorganic reagents - FISher certified NaOH and HCIwere used for pH modification. ACS reagent grade NaCI, aproduCt of Amend Drug and Chemical Co., was used foradjusting the ionic Strength.

Water - Triple distilled water (TOW) of specific con-ductivity of 10-6 mho was used in this investigation.Experimental

Materials

Techniques

Adsorption tests - 0.4 g (0.0014 oz) of material was~uilibrated with 8 an' (0.5 cu in.) of "X 10-2 kmol/m'NaCI ~lution in a screw cap glaM vial for four hours byshaking at the desired temperature in a wrist action shaker.After four hours of equilibration, pH was adjusted usingNaOH or HCI and the suspension was further equilibratedfor two hours. Required amount of polymer ~lution wasthen added and the suspension was agitated for an addi-tional12 hours before adding the surfactant ~lution. Mteradding the surfactant, two more hours of agitation wasrequired to reach the equilibrium. After equilibration, pHof the smpension was measured using a thin glaM electrodeattached to a digital Corning pH meter (Model 125). Thecontents of the vial were transferred to a centrifuge tube andwere centrifuged in a IEC Model B20-A centrifuge at15,(XX) rpm for 10 minutes. Residual amount of the polymerand the sulfonate were detennined in the supernatant andadSAJrption densities were calculated using the respectivecalibration curves.

Reproducibility of the adSAJrption measurements at lowerresidual polymer concentrations ( < 1(xx) mgikg, or 0.07 ozper lb) was found to be within 4%. Accuracy of measure-ments generally was better in the lower polymer COnceDtra-tion range.

Analytical - the amount of Cl. labeled polymers was

determined using a Beckman M<Kle1 1.5 l00c spectro-photometer .

Hematite - synthetic hematite, (99.2-99.5% Fe2<' 'S.:was obtained from Pfizer Inc., and its particle size ~ "reported to be 99% 1e8 than 5 JAm (2500 mesh). Surfacearea of the sample using nitrogen as the adSl>rbate wasdetermined by the Quantasorb to be 8.7 m!/g (3.3 cu ftper oz).

Polymers - C14 tagged nonionic and ionic polyacryla-mides were synthesized using radiation induced hetero-genous polymerization technique. The synthesis procedurewas similar to that recommended by Wada, Sekiya. andMachi, 1975, 1976). An overall monomeric concentration of2.5 kmol/m5 was maintained in an acetone (45%)-water(55%) solvent at pH = 7.5. Polymerization was achievedusing a Co60 source at a dose rate of about 1 k Rad/h.Additional details of the synthesis and characterizationtechniques have been given ebewhere (Moudgil, 1981).

Anionic copolymers (P AMS) were synthesized using 2-acryiamido-2methyl propane sulfonic acid (AMPS) aproduct of Lubrizol Corp as the comonomer .

Cationic polyacrylamides (P AMD) were synthesized usingdimethyl-aminopropyl methacryiamide (DMAPMA) as thecomonomer. This reagent was received in the stabilizedfann from the Jefferson Chemical Co. The hydroquinoneinhibitor was removed by passing a 50:50 aqueous solutionof this reagent through an activated carbon column. Theaqueous solution obtained was used immediately after theinhibitor removal stage.

VISCosity average molecular weight of the polymers was~ted using the Mark-Houwink relation [,,] = K[M]a,

SocjeIy 01 Mining Engi~.. 01 AlME Tra.-ctio.. Vol 27&-1G1

'" -

Results and ~ondIA

aL

In:

~islra

SUIairin

eq,m(tIxcorest.

sptmttbtcnoth(thftio'USI

Table 2 - Chromatographic Separation of Nonlonlc

Polyacrylamide (PAM) During Adsorption on HematitePrP1iminary studies were conducted to evaluate the effectof parameters such as agitation intensity and time, whichfannulate the equilibrium adsorption test procedures. Also,the reversibility of adsorption of the polyacrylamide typepolymers on hematite was investigated since knowing the ex-tent and rate of desorption helps in understanding themechanism of the polymer adsorption p~.

NaCl Conc. = 3 x 10-2 kmol/m3pH = 3.5

Molecul81 Wi. Belore Adsorpllon = 2.36 x 106MolecularWt.Alt8rAdsorptlon = 2.22 x 106

Effect of Agitation on Polymer Adsorption PropertiesDesorption-Reversibility of Polymer Adsorption

To detennine if shaking the polymer ~lution duringequilibriwn adsotption tests modifies its adsotption behaviorin any way, non-equilibrium adsorption tests involving onehour of agitation were conducted. One of the polymer~lutions ~ in the adsotption test was shaken in a wristaction shaker for 12 hours (duration of the equilibriumadsotption tests) and the other was kept on a bench top.Adsotption results obtained after one hour of agitation arepresented in Table 1. Also, to detennine if the two agitationconditions had resulted in any substantial changes in themolecular weight distribution of the polymers, intrinsicviscosity of the supernatant was measured using a capillaryviscometer. The estimated molecular weights are also~ted in Table 1. Experimental data indicates thatagitation during the adsotption tests did not ca~ changesin the structure of the polymer molecules to affect theiradsotption behavior on hematite.

It is possible to obtain some knowledge of the strength ofthe bond fofDled between the polymer speci~ and thea~rbent by determining the extent of desorption or thereversibility of the polymer a~rption p~. The type ofsolvent has been known to affect the rate and amount ofdesorbed polymer to a considerable extent. Desorptionnonnally is slow and incomplete when the polymer isa~rbed from dilute solutions, whereas, from concentratedsolutions, it is rapid and frequently complete (Lipatov andSergeeva, 1971). This could be the result of differenc~ inthe stability of the a~rbed polymer from dilute andconcentrated solutions. In the fonner case, optimumnumber of bond fofDlations can occur between the polymermoIecu1~ and the a~rbent. However, in the case ofa~rption from concentrated solutions, less than optimumnumber of bond fonnations are possible due to "crowdingin" by other polymer molecu1~ on the a~rbent surface.

An attempt was made in this investigation to desorb thepolymer from the hematite surface by decreasing the bulkconcentration (by replacing part of the solution abovesettled soli~ with salt solution or roW) and shaking thesuspension in a wrist action shaker for an additional 12hours. It was expected that desorption of the a~rbedpolymer molecu1~ would increase the "new" residualpolymer concentration. A decrease in the residual concen-tration, on the other hand, would indicate continueda~rption. Table S results make it clear that desorption ofthe polymers upon dilution, irrespective of their ionicnature, did not occur under these experimental conditions.

~F~£01]a.'.thf:..aJnW3!

p»del

~I

Uk

ofFLar

mo

var

prothf:

an<:.

can

invl

viol;

~isoti

\'en

(

tYJ'(buiS). .

pH'

grol11.5

neg;iDa'

~A(

to iT.

daD<

Table 1 - Effect of Mechanical Agitation on Adsorption ofNonionic Polyacrylamide (PAM) on Hematite

Under Neutral pH Conditions

Inltl8' Polymer Cone.mole2000~

2000250

NaCI Con. = 3 x 10-2 Kmol/m3pH~6.8

Agitation Time = 12 HoursAdsorption Time ~ 1 Hour

AoItatlan Ad8orpllan Dan8lty M. Wl 01 PolymerIn Supematant

No 17.4 2.36 x 106No 6.0 2.38 x 106Ves 17.5 2.32 x 106Vea 6.7 2.25 x 106

Table 3 - Desorption of Polymers

Chromatographic SeparaHon of PolymerDuring Adsorption

It has been reponed that displacement of initiallya~rbed low molecular weight fraction by higher molecularweight fraction could occur if the molecular weight distribu-tion of the polymer is broad (Felter. Moyer. and Ray. 1969;Kothoff and Gutmacher. 1952). To deteImine if there wasany preferential a~rption of higher molecular weightfractions during the present investigation. molecular weightof the polymer before and after the equilibrium a~rptiontest was estimated. Table 2 results show that molecularweight estimates of the polymer before and after completionof an a~rption test were within :t 10%. the accuracy limitof the intrinsic viscosity technique used for molecular weight~tion. This indicated that either the molecular weightdiStribution of the polyacrylamide used was not broadenough to result in any measurable chromatographicseparation during the adsorption proceM. or that the presentsystem was not capable of causing any such separation.

Polymer pH

NaCl Cone. = 3 x 10-2Adsorption Time = 12 Hours'Desorption Time' = 12 Hours

Initial Polymer Cone. = 1000 mg/kg

Realdua' Polymer Conc. A..ldual Polymer Cone. afteraftar 12 Hra. of Adaooption 12 H... of 'Deaooption,'

mg/kg mg/kgEstimated experimental

671 251 248723 271 271718 269 261470 176 181e22 233 222835 2a 2:M)882 28 245707 - 278

PAM ~.278..e..3.12e.-8.538.8.78

PAMS

PAMO

It has been pointed out that "i~rsibility" observed inthe case of polymer a~rption is not true iITeversibility, andadsorption results can satisfactorily be explained on the basisof polydispersity of the polymers empl~ in such studies(Cohen Stuan, Scheutijens, and 1'1eer, 1980; Koopal, 1981;and mady. Lyklema. and Heer. 1982). It was noted that .'--T ctIo.. Vol. 278

Ioci8tJSociety of Mining Engine... of AIME

during ~ dilution prtx:s to ~ the desorption, thesurface to volume ratio (S/V) is decreased. That can have amajor influence on the polymer adlOrption. Consequendy,this might lead to the ~ conclusion that the polymeris ~bly adsorbed. In the present .ooy. OOwewer, S/Vratio was maintained constant becaUR a given volume of thesupernatant after adlOrption was replaced with an equalamount of the liquid without any polymeric speci~ presentin it. AIao, in the CaR of poiydisperled polymen underequilibrium conditions, it was reported that the "higher"molecular weight fractions are preferentially adlOrbed andthe residual polymer in the bulk solution should euentiallyconsist of "lower" molecular weight fractions. Once again,estimating the molecular weight of the residual polymericspecies in the present study, using the intrinsic viscositymeasurements, did not reveal any significant difference inthe molecular weights calculated before or after the adsorp.tion tests. In other words, no chromatographic separation ofthe polymeric speci~ was obrerved. It should be noted thatthere are no reports that prove the reversibility of adsorp-tion, even in the case of monodispened polymeric systemsusing a ~lvent of similar thennodynamic character .

fX poiyaaylamida on hematite cxcun by ~ bondingof the amide group to the h~ donating surface sites.which will in~ with a decreaR in pH (Linke and Booth.1959; Gebhardt and Fuemenau, 1982). At pH valuehigher than 8.1, DOt only the hematite surface is negati~y~. but the polymer itself will be anionic in nature.FJectrc.tatic charge repuJsion at such pH values, therefore.will become significant. resulting in reduced adlOrption.

In the CaR of anionic polyacrylamide with a sulfonatefunctional group (PAMS). its adsorption behavior onhematite as a function of pH is similar to PAM (fig. 2).Anionicity of the polymer at pH above 8.1 and therefore.~tic charge repuJsion. will be ~ more significantinthis~.

Adsorption of the cationic poiyaaylamide (PAMD) OIlhematite at three pH I~ls is plotted in fig. S. At pH 2.7and 7.2, hematite will exhibit a net positive charge. thecharge density being higher at the lower pH value.Howe\oer. at both the8e pH levels, tM polymer chargecharacteristics will reInain unchanged. Therefore. beca~of the electrostatic repuJsion. polymer adlOrption onhematite by h~ bonding will decrease. The reductionin the polymer adlOrption will be more pronounced at pH2.7 than at pH 7.2.Adsorption of Polymers on Hematite

IoS..,._,.s ~."

-0- ).4 t 0.1.0. 6'.t 01-0. 1O.0.t OJ

:~- tt.} 10

- .1~~ 11

I.. .

~ ., .i 10

0 I..t 5

4

20 '

0,- I I I I I I

ZOO 400 tOO ~ I~ ~ MOO ~ ~~

RESIDUAL C~C[HTbTIOH 1.".,1

Fig. 1 - Equilibrium adsorption Isotherms of nonionlcpolyacrylamide (PAM) on hematite

';.

r.....i~

8

i0c

pH ED_a: Ad.1orption of acryiamide baaed IKJmO- andcopolymen on hematite under different pH conditions ispresented in Figs. 1 to 5. M~ of the adsorption isothennsaxe characterized by a higher slope at low concentrations~~ by a slower uptake of the polymer at higherconcentrati~. A plateau. however, is DOt reached withind1e polymer concentration range tested. It was determinedalso that. up to a cenain initial polymer concentration,ahnost all of the polymer was adlOrbed on the solid. and itwas only at higher concentratiom that a partitioning of thepolymer between the surface and d1e bulk solution wasdetected. This type of adsorption behavior is characteristicof monolayer type of adaorption for nonpolymeric materialand can be dcscribed by the Langmuir equation. In the CaRof polymer adsorption. however. any fit of the data to theLangmuir equation is probably fortUitous since polymermolecules can asume different conformations and exhibitvarying degree of attacluncnt to the surface as adsorptionPlo>g1~. Moreover. adsorption of poiyacrylamides used ind1e p~t study was found to be ineYersible upon dilution.and lateral interactions betWeen adsorbed polymer molecul~cannot be nJJed out. Thus, two of the basic a8umptionsimoIved in the derivation of the Langmuir equation axeviolated. Although adsorption iaotherms generated in thepraent study were similar in shape to the Langmuirisotherm, they did not satisfy the criterion of a linear C/ A\'eISUS C relation.

One C(XDInon feature of the adaorption of polyacrylamidetype poIymen on hematite is that adaorption is l~ underbasic pH conditions than at acidic or neutral pH (Fig. 1 to5). This can be attributed to the polymer hydrolysis abovepH 8 resulting in the formation of carboxylic acid fuoctionalgroups in all three types of poiymen. For example, at pH11.5. hematite is negatively chargm aM introduction of anynegative charges on the polymer backbone will result in anincrease in the electrostatic repulsion, thus reducing the

polymer adaorption.Adsorption of nonionic polyacrylamide (PAM) was foond

to inaeue with decreasing pH (Fig. 1). This is in accor.dance with the mechanism p~ earlier that adaorption

Fig. 2 - Equilibrium adsorption Isotherms of anionic poly,acrylamlde (PAMS) on hematite

Society of Mining EngI.-IS of AlME Tr:.":aGtIG.~ Vol ~1-

Adsorption of nonionic polyacrylamide (PAM) on hema-tite at 25°, 45°, and 75°C (77°, 115°, and 167°F) is plottedin fig. 6. Similar adsorption behavior was observed at 25°Cand 45°C (77° and 115°F). However, at 75°C (167°F), atrend towards higher adsorption was obtained. Such behav-ior L, indicative of significant influence of the entropiceffects on the adsorption of procea.

.. - 7.. toO.'~~3

-0-'-0- 3010-2-0- ,..--- ..,

--" 0~~~~

IX

.r....Vtz

~

~;:Q.

~tit0C

24

m

1.

12

.

,.~.

1""- . I . . . . . . . .

I - - t- ~ ~

RESIDUAL CONCENTRATION (m9/~91

Fig. 4 - Effect of salt addition on nonionlc polyacrylamide(PAM) adsorption on hematite under neutral pH conditions

24

- 22... 20f- II..= 16.;:: Ci 10 //

~ :1

IiI,.. ...

...

011' 70tO1

MOO CONCENTAATIOI

.01/.'03 . 10-1

u. I I I I , I I I , I0 200 400 600 ~ 1000 ROO 1400 .00 1800 2000

~ES~UAL COfiCENTRATION .."",Fig. 5 - Effect of salt addition on anionic polyacrylamide(PAMS) ad$orptlon on hematite under neutral pH conditions

21 I' 5 .~_/.,sN8CI,"'~5t0224 - Z~"C

22 -<J- 45"C~ ~ '~'Cr 20- II.... ..

~a 14~5 12

0: 10

~-QC

4

RE$IOUAL CONCENTRATION I.../kt'

Fig. 6 - Effect of temperature on adsorption of nonionicpolyacrylamide (PAM) on hematite

At pH 10.9, hydrolysis of me cationic polyacrylamide willinduce negative charges (RCOO-) on the polymer back-bone. This results in el~tic repulsion between similarlycharged polymer and hematite panicl~. The negativechaIge repulsion at pH 10.9 can be expected to be moresignificant than the positive chaIge repulsion at pH 2. 7 and7.2, since the amount of cationic amine functional groups (Smol %) is less than the amide groups in the polymer, whichwill hydrolyze to RCOO-. It is therefore expected that thepolymer adsorption will be lower at pH 10.9 than at me twolower pH levels (Fig. S).

Ionic StTengeh Effect: Ionic strength changes can affectpolymer adsorption through modifying the solvent power ofme medium. and increased competition between coun-terlons and polymer segments for adsorption at the solidiliquid interface. If the polymer is more soluble at higherionic sttength. a decrease in adsorption would be expected.The increased competition between added ionic speci~ andpolymer molecul~ would also lead to reduced adsorption. Athird factor to be considered for polyeiectrolytes is theP<8ible modifications in polymer conformation as a resultof the added indifferent electrolyte (Muller. Laine. andFenyo. 1979). El~tic repulsion between chargedfunctional groups will be minimal in the presence of salt.permitting increased coiling of the polymer chain. Thisshould lead to a reduction in area per molecule. and. mus,an increase in adsorption.

Adsorption of PAM (nonionic polyacrylamide) on hema-tite at different concentrations of NaCI is plotted in Fig. 4.There is no significant effect of varying me salt concentra-tion from 0 to 5.1 bnol/m5 NaCI on me adsorption of PAMon hematite. These results indicate that there are nocharged sites (due to hydrolysis) associated wim the polymerbackbone. Considering that adsorption of polyacrylamideon oxide mineraJs is by hydrogen bonding, no significanteffect of ionic strength is expected. Adsorption of P AMS(anionic poiyacrylamide-a pol:)'electrolyte) on hematite (Fig.5) was deteImined to be higher at higher salt concentration.This suggests that changes in polymer conformation mightinfluence the adsorption of P AMS more than the increaseda.npetition between me counterlons.

Tem~uTe Effect: Variations in temperature canaffect polymer adsorption through changes in solvent powerof the medium. and adsorption of the solvent molecul~a.npeting wim the polymer. Both factors have similar rol~in monofunctional solute adsorption, except that forpolymers, entropic effects due to adsorption of the solventmolecul~ competing wim the polymer can be considerable.

1834-Tranl8ctiona VOl. 276 Society of Mining Engi-. of AlME

Conclusi~

Based on the above mscumon, the following conclusionsregarding the adsorption of polyacrylamide type polymerson hematite are reached.

Adsorption of the ionic and nonionic polyaaylamides onhematite is governed primarily by hydrogen bondingbetween the surface oxygen and the oxygen or nitrogen ondie polymer. Dec~tic interactions between the polymerfunctional groups and the hematite surface play only aRCOndary role in the p~t system beca~ of the relativelylow degree of functionality of the copolymers.

The relatively small effect of salt addition on the adsorp-tion of anionic POlyaaylamide (P AMS) on hematite and thenegligible effect of ionic strength on the adsorption ofnonionic polyacrylamide (PAM) on hematite indicates thateither the changes in parameters such as .,lvent power ofthe medium and conformation of the polymer are notsignificant, or, their effects are mutually cancelled out.

The effect of temperature on the adsorption of poly-acrylamide on hematite was not found to be significant,leading to the conclusion that entropic, rather than enthal-pic changes, playa major role in the adsorption process.

Desorption tests indicated that, im:spective of thepolymer charge, polymer adsorption on hematite upondilution can be considered to be irreversible under presentexperimental conditions. .

Acknowledgments

The authoR would like to thank Dr. K.P. Ananthapad-manabhan for helpful discussions and Regina Gorelik forhelp in experimentation. Financial support of the NationalScience Foundation (Grant No. DAR-79-09295) and IncoIoc.. is acknowledged. B.M. Moudgil also thanks theOccidental Research Corp. for granting educational leaveto him at Columbia Univenity .

Referen~

Felter, R.E., Moyer, R.S., and Ray, I.N., 1969,J. Polym.Sa.., Vol. B7, p. 5~.Gebhardt, J.E., and Fuentenau, D. W., 1982, "The Adsorp-tion of Polymers on Oxides," Interfacial Phenomena inMineral PrOcessing, Yarar, B., and Spottsiwood, D.J., eds.,The Engineering Foundation, New York.Higuchi, M., and Senju, R., 1972, PolymJ., Vol. 3, p. 370.roady, V., Lykiema, J., and fleer, GJ., 1982,J. ColloidInterface Sa.., Vol. 87, p. 395.Huggins, M.L., 1958, "Physical Chemistry of High Poly"mers," John Wiley and Sons, New York.Koopai, L.K., 1981, J. Colloid Interface Sci., Vol. 83,p.116.~, I.M., and Gutmacher, R.G., 1952,J. Ph)lS'. Chern.,Vol. 56, p. 740.Kulicke, W., and Klein, J., 1978, Angeucndte Makromol.Chem., Vol. 69, p. 189. Quoted in Hollander, A., 1979, "AStudy of the Interactions of Poly (Acrylamide) and ItsCopolymen with Kaolinite," M.S. thesis, Columbia Univer-sity, New York, NY.Linke, W.F., and Booth, R.B., 1959, Tnms., AIME, Vol.217, p. 364.Lipatov, Yu.S., and Sergeeva, L.M., 1971, "Adsorption ofPolymen, " John Wiley and Sons, New York.

Moens, J., and Smets, G., 1957,J. Pol)'m. Sa.., Vol. 29,p.931.Moudgil, B.M., 1981, "The Role of Polymer-SurfactantInteractions in Interfacial Pl-uO:~," Eng. Sc. D. thesis,Columbia University, New Yark, NY .Muller, G., Laine, J.P., and Fenyo, J.C., 1979,J. Poiym.Sci. (Polym. Chem.), Vol. 17, p. 659.Onoda, G.Y., Jr. and deBruyn, P.L., 1966, Surface Sci.,Vol. 4, p. 48.Parks, G.A., and deBruyn, P.L., 1962,J. Ph)lS'. Chem.,Vol. 66, p. 967.Pinner, S.H., 1953,J. Polym. Sci., Vol. 10, p. 376."Polymer Handbook," 1974, Brandrup, J., and Immergut.E.H., eds., Second Edn., John Wiley and Sons, New York.Smets, G., and Hesbain, A.M., 1959,J. polym. Sa.., Vol.4O,p.217.Somasundaran, P., and Moudgil, B.M., 1982, "Effect ofPolymer-Surfactant Interactions on polymer Solution Prop-erties," Macromolecular Solutions, Seymour, R.B., andStahl, G.A., eds., Pargamon Press, New York.Wada, T., Sekiya, H., and Machi, S., 1975. J. Appl.Polym. Sci. (Chem.), Vol. 13, p. 2375.Wada, T., Sekiya. H.. and Machi, S., 1976, J. Appl.Pol)'m. Sci., Vol. 20, p. 3233.

Ahmed, S.M., 1975, "F1ectrochemical Properties of theOxide Mineral Interface in Relation to F1otation," A dwncesin Interfad4l Phenomena of Particulate / Solution / GasSystems: Applications to Rotation Research, Som~a-'--~~ran,P., and Grieves, R.B., eds., AIChE, SympclSium Series Vol.71, No. 150, AIChE, New York.

Atkinson, R.J., P~, A.M., and Quirk. J.P., 1967, J.Phys. Chern., Vol. 71, p. 550.

Cohen Stuart, M.A., Scheutijens, J.M.H.M., and F1eer,G.J., 1980,"J. Poly. Sa. (Polym Phys. Edn), Vol. 18, p. 559.

Tran88Clion8 Vol 27&-1835Soci8ty of Mining Engineers 01 AlME

~~~~.~--

TECHNICAL NOTE

C.H. Baab and LS. Koibert

mtroduction

Types of Nuclear Instnunentationin Coal Preparation Plants

In JWle 1981. the Electric Power Research Institute(EPRI) issued a report tided "Conttol Systems in CoalPreparation Plants" to determine the status of instrwnenta-tion and automation systems in the coal preparationindustry. EPRI concluded that the conttol systems presendybeing employed in the US coal preparation plants "arerelatively primitive compared with th~ in other mineralprocessing industries; and analytical instrumentation isvirtually non -emtent. ,.

With tighter specifications on the quality of steam coal,interest is increasing in more "phisticated conttol systems.One type of device that is CWTendy a..cring the preparationplant operator optimize and conttol the coal cleaningprocess is nuclear based instrumentation. Nuclear instru-mentation includes methods for measuring weight, level,and d~ty as well as for perrOm1ing elemental analysjson coal.

There are four types of nuclear instrumentation preseDdybeing employed in coal preparation plants. They are:

. Belt Weigh Scales

. Point Level Switches

. Density Gauges

. Coal Anal}lSis Equipment.Each type of instrumentation has a number of potential

applications within the plant. What follows is a discussion ofthree different types of devices and their applications.

Weigh Scales

Major Features and Benefits of NuclearMeasurement and Analysis Systems

There are many features and benefits of nuclear instru-mentation. The major benefit is that nuclear instrWnenta-tion is non-contacting. The measuring heads do not contactthe process materials. This enables measurements of corro-sive, abrasive, or high temperature materiaL, with no 1<8 ofperformance, reliability, or useful life. AJso, there are nomoving parts to wear or be fouled by dust or ~on.

Installation is simple and inexpensive. For instance,~ty gauges are simply clamped onto ~ pipe nlnS.

Weigh scales and level switches usually require nothingmore than welding brackets to customen' equipment. Onceinstalled, routine maintenance can be performed withoutinterrupting the process. Thus, nuclear instnunentation ishighly reliable and easily installed and maintained. Theoperating life of the instrumentation is virtually unlimited.

Principle of Operation

In coal preparation plants, weighing instruments are usedfor measuring material throughput, totalized weight. andbatch weight on conveyor belts and other transport equip-ment. They are aJso used on drag chain conveyon. vibratingconveyon. and screw conveyors. Weighing is important forreceiving the run-of-mine coal, controlling mag flow,scheduling production, and loading and shipping thewashed coal to market.

There are two basic types of weighing instruments:gravimetric and nongravimetric devices. Gravimetric devicesare units which measure the pull of gravity on the materialbeing weighed. Nongravimetric devices are not dependenton the force of gravity on the sample. Nuclear weigh scalesare the most prevalent nongravimetric device used.

The nuclear weigh scale works on the same basic principleas other nuclear gauges. The greater the quantity ofmaterial placed in the path of a radiation beam, the moreradiation will be absorbed by the material. In weigh scales,Cesium 1~7 source u generally placed above the movingstream of material so that the radiation is directed throughthe material. An ion chamber radiation detector is on theopposite side of the material. The radiation paging to thedetector generates a signal which u invenely proportional tothe weight of the material present. This signal is amplifiedand integrated with the speed of the proces flow yieldingthe weight. The.weight is indicated and can be totalized forthroughput. batch control, etc.

In conclusion, radiation weighing methods have becomewell established in the coal preparation industry throughoutthe world. As mentioned before, this device finds its ~

Nuclear belt weigh scales, point level switches and densitygauges all operate on the gamma-ray transmission principle.A beam of gamma radiation from a radioOOtopic source isprojected through the procea material either on a belt, in avessel, or in a pipe. Opposite the source is a radiationdetector w~ electrical output is proportional to theintensity of the radiation it absorbs.

Ganuna radiation is part of the same electromagneticspectrum that encom~ light, infrared, and x-rays. Inmany respects, gamma radiation behaves much like light.However, it is far more penetrating and can be transmittedthrough considerable thickness of materials that are opaqueto visible light, e.g., several inches of steel.

C.H. Baab and L.S. Kolbert, regional sales manager, are withTexas Nuclear Corp., Austin, TX. SME preprlnt 83-518, SME-AIME Annual Meeting, Atlanta, GA, March 1983. ManuscriptMarch 1983.

185-TralW8ClIon8 Vol. 276 Society of Mining Engi.-s of AlME