JOURNAL 01' cou.om AND INTERP ACB SCIENCE 177, 283 - 287 (1996)ARTICLB NO. 0033

Role of Polymer Conformation in Interparticle-BridgingDominated Flocculation

XIANG Yu AND P. SOMASUNDARANI

Langmuir Center for Colloids and Interfaces, Henry Krumb School of Mines, Columbia University, New York. New York 10027

Received May 30. 1995; accepted July 25. 1995

mers, both of which are ascertained to be main controllingfactors in flocculation with polymeric flocculants (4).

Our previous studies have discussed the importance ofpolymer confonnation in flocculation of alumina suspensionwid1 polyacrylic acid and a method using pH shifting tomanipulate d1e confonnation to achieve enhanced floccula-tion or dispersion as desired (5). Recently, we have obtainedmarkedly better flocculation using a combination of poly-mers (6). The aim of this work is to elucidate d1e mecha-nisms involved in such flocculation. The electrostatic forcewas isolated by comparing d1e dual polymer perfonnance atone pH wid1 d1at of single polymer at a pH where d1e zetapotential is identical.

EXPERIMENTAL

Materials

Linde A alumina powder used in this study was purchasedfrom Union Carbide Corp. and had an average size of 0.3.urn and BET surface area of -14 m2/g. Polyacrylic acid(PAA) of molecular weight lOO,<XX> and polydiallyl di-methyl ammonium chloride (PDADMAC) of molecularweight 240,<XX> were purchased from Polysciences Inc. ACScertified grade hydrochloric acid (HC1) and sodium hydrox-ide (NaOH) solution from Fisher Scientific Inc. were usedas pH modifiers. Ionic strength was kept constant by using0.03 mol/liter NaCl solution in all the experiments.

Flocculation of alumina with a dual polymer combination (po-Iydiallyl dimethyl ammonium chloride and polyacrylic acid) iscompared in this study with individual polymers. Zeta potentialand conformational characteristics of the system are studied inorder to determine the role of these properties in determiningflocculation. The dual polymer system is shown to yield markedlybetter flocculation than either single polymer. Analysis of floccula-tion results along with the zeta potential measurements undercorresponding conditions showed polydiallyl dimethyl ammoniumchloride flocculation of alumina to be mainly electrostatic whilepolyacrylic acid flocculation to be polymer bridging at high pHand charge neutralization at low pH. The role of polymer confor-mation was found to be very important upon isolating the electro-static effect by comparing the flocculation results under constantzeta potential. Polymer conformation in solution as well as at thesolid-liquid interface can be altered by changing solution condi-tions and complexation with another polymer. It was shown thatby manipulating polymer conformation, the flocculation can begreatly enhanced even at very low polymer dosage. In the specificcase studied, the stretched and strongly adsorbed polymer wasfound to give excellent flocculation. ~ 1996 Academic P Inc.

Key Words: flocculation; polymer combinations; polymer confor-mation; polymer complexation; zeta potential; alumina; polya-crylic acid; polydiallyl dimethyl ammonium chloride.

INTRODUCTION

Methods

Polymers are increasingly used as flocculants in solid-liquid separation and sludge dewatering processes (1). Inmany cases, use of two or more polymers has been foundto enhance the flocculation markedly. However, while theeffect ~f charge density, chemical structure, and dosage ofpolymer,s has been studied in the past, not much attentionhas been'~ven to understanding the mechanisms of dual ormultipoly~~occulation (2,3). Polymer-polymer interac-tion in solution or at the solid-liquid interface depends onthe nature of polymers and solid surface, solution chemistry,and the interaction can profoundly change the charge proper-ties of polymers and more vitally the conformation of poly-

Flocculation tests. For each test, a 5 g sample of aluminawas stirred in 190 ml of 0.03 mol/liter NaCl in a 250 mlbeaker with a magnetic stirrer for 30 min. The pH of thesuspension was then adjusted to the desired values and diesuspension was further conditioned for 30 min. The beakerwas then fitted with four !-in. wide baffle plates and a I-in.diameter propeller for better mixing and stirred with thedesired amount of polymer stock solution for 1 min at 600rpm. In the case of the dual polymer systems, the secondpolymer was added 1 min after stirring with the first polymer.The sample was stirred for an additional 4 min at 300 rpm.I To whom correspondence should be addressed.

283 0021-91971% $12.00Copyright Q 1996 by Academic Pte.'IS, Inc.

All rights of ~tion in any form reserved.

284 YU AND SOMASUNDARAN

obtained for flocculation of alumina at pH 10.5 with polydi-ally dimethyl ammonia chloride are shown in Fig. 2 alongwith the zeta potential. It can be seen that the optimumflocculation (maximum settling) was obtained when the zetapotential was close to zero, which suggests that the floccula-tion in this case is controlled primarily by electrostaticforces. Adsorption results also show the positively chargedpolymer to adsorb completely on the negatively chargedalumina at pH 10.5 in the concentration range studied. It isnoted that the settling obtained with polymer at IEP isslightly different from that obtained without polymer at IEP.This suggests that apart from charge neutralization polymerbridging also plays a role in the flocculation. But, it is sur-prising that this bridging effect is minor even though themolecular weight of the polymer is high enough to cause

good bridging.

Flocculation with Combined Mechanism

Aocculation of alumina with polyacrylic acid is shown inFig. 3 as a function of pH along with the zeta potential.It can be seen that with the addition of P AA, maximumflocculation is obtained still close to pH 8 in comparison topH 8.5 without P AA, which is close to the IEP of aluminaafter P AA adsorption. This suggests that the electrostaticeffect is the governing factor in this flocculation also. How-ever, it should be noted that the alumina suspension is floc-culated with PAA even when both the polymer and the parti-cles are similarly negatively charged at high pH. It has beensuggested (6) that polyacrylic acid can adsorb on aluminaeither by electrostatic force at pH lower than IEP or viahydrogen bonding at high pH. The flocculation obtained hereat pH below IEP is attributed to electrostatic adsorption ofP AA resulting in interparticle bridging and charge neutral-ization. When pH is above the IEP, the polymer molecules

The slurry was then transferred into a 250 ml graduatedcylinder for settling rate measurement.

Zeta potential measurements. The zeta potential wasmeasured using a laser zee meter (Pen Kern Model 501).After the flocculation test was completed, flocs were ultra-sonicated for 30 s to break them to small particles and dilutedwith supernatant in order to obtain a suitable dilute sample.

Polymer adsorption tests. Polymer adsorption on alu-mina was estimated by determining its depletion from thesolution. The samples were treated in a fashion similar tothat used for the flocculation tests and supernatants werethen separated from sediments by centrifugation at 3000 rpmfor 15 min. Polymer concentration was measured using aDohrmann total organic carbon analyzer. Blank tests withpolymer or polymer mixture solutions at the same concentra-tions without alumina were carried out to ensure that adsorp-tion results were not affected by any precipitation.

.0 40

RFSUL 1'8 AND DISCUSSION

0.8Electrostatic Flocculation ~

8-0-

:3'

!0

~

!

14f~~

,§itn

Flocculation is considered to be governed by two majormechanisms, the electrical double layer compression orcharge neutralization and interparticle polymer bridging.However, the latter has not been as well understood as theformer mainly because of the lack of a reliable technique todetermine polymer conformation. In order to understand thebridging effect, it is necessary to isolate electrostatic forces.Alumina flocculation without polymer was done as a func-tion of pH as a reference for comparison since it is dependentonly on the electrostatic force between particles (see Fig.I ). It can be seen that flocculation results correlate well withthe zeta potential of alumina particles. The maximum settlingoccurs at the isoelectric point (IEP) of alumina. Results

0

.20

0.6

0.4

0.2

0.0 -40

0 10 20 30 40 SO

Conc. ofPDADMAC, ppm

FIG. 2. flocculation of alumina with polydiallyl dimethyl ammoniumchloride at pH 10.5 (0 Settling rate, 'V Zeta potential).

285POLYMER CONFORMA nON IN FLOCCULA nON

402.2 ColIC. ofPDADMAC, P..-0.1 I 10 100 1000

40 .. .., ... . .. .., ... ..., .. ...2 20

30

0

-»

118..~16~=ISU 1.4tI)

0-401.2

20>e 10-aOJ 00A.; -10N

-20

.6() -306 8 9 to

pH-40 ~- I I - I I ,- - .. " - - ..

11 10 9 8 7 6 S

pHFIG. 3. Flocculation of alumina suspension with 20 ppm polyacrylic

acid as a function of pH (0 Settling rate, \l Zeta potential).FIG. S. Comparison of the zeta potential of alumina particles as a

function of pH (l:1) and concentration of PDADMAC (0).adsorb via hydrogen bonding and inducing interparticlebridging causing flocculation even when the zeta potentialof alumina particles is high.

Flocculation of Alumina Dispersion with Dual Polymers

Flocculation results obtained with the sequential additionof JK>lydiallyl dimethyl ammonium chloride and polyacrylicacid are given in Fig. 4. It can be seen that the flocculationis markedly enhanced by the subsequential addition of P AAafter the cationic JK>lymer preadsorption. This is attributedto the efficient interparticle bridging by polyacrylic acidwhich was added subsequent to the adsorption of the oppo-sitely charged PDADMAC molecules. Zeta potential mea-surements showed that P AA addition to the system does notaffect the surface charge of alumina particles significantly.

This is possibly due to the low dosage of polyacrylic acidused in comparison to polydiallyl dimethyl ammonium chlo-ride. Adsorption results show the PDADMAC to stronglyinteract with alumina with no polymer left in the supernatantin the concentration range studied. P AA was also shown tocompletely adsorb on the alumina surface for all experi-ments.

Unlike in the case of single polymer systems, there isno longer a definite correlation between zeta potential andflocculation, indicating that electrostatic force is no longerthe only governing factor for this flocculation.

Comparison of Flocculation with Singleand Dual Polymer Systems

In order to isolate the electrostatic effect, the zeta potentialof alumina is plotted together in Fig. 5 as a function of pHand as a function of PDADMAC concentration. By choosingthe proper scale, both curves are made to overlap, so thatzeta potentials at a given concentration of PDADMAC areidentical to that at a specific pH. Aocculation of aluminawith PAA at different pH conditions and PDADMAC ad-sorptions are compared in Fig. 6 using the same scale as inFig. 5. Since the zeta potential of particles along the x axison both curves is identical, the property that could haveyielded different is proposed to be interparticle bridging thatis dependent upon the polymer conformation at interface. Itcan be clearly seen that the flocculation response with dualpolymers is better than that obtained with either polymeralone with the difference being particularly marked at lowpH values.

3

2.S

j 2

S~ 1.5

f 1Con

0.5

0 I . . . . I . . . I . . . . I . . . .

0 10 20 30 40 SO

COlIC. ofPDADMAC, ppm

FIG. 4. flocculation of alumina suspension widt PDADMAC and PM(0 PDADMAC alone, TV PDADMAC + 2 ppm PAA, 0 PDADMAC +5 ppm PAA).

Change of PAA Conformation in Solution

Polyacrylic acid, with carboxyl groups along the back-bone, can be expected to possess different conformations at

~...'i0~

~N

286 YU AND SOMASUNDARAN

Conc. oCPDADMAC. p.-n0.1 1 10 100 1000

0.55 .. . . .. ." ... . ." ."., '" ... . . I."]

0.5

0.4S

0.4

O.JS

0.3

0.2S

~~~

:'3.0u

0.2

II 10 9 8

pH

1 ,6

FIG. 6. Comparison of ftocculation response as a function of pH (b.)and concentration of PDADMAC (0).

0.1 S ' . . . I I . . . I . . . . I . . . . I . . . .

II 10 9 8 7 6 S

pH

FIG. 8. Conformation change of PAA at the solid-liquid interface asa function of pH (~) and concentration of PDADMAC (0).

different pH due to the pH dependent ionization ratio ofCOOH groups and interactions of these groups with otherions in the system and functional groups on other polymers.The fluorescence spectroscopic technique was used here todetenmne conformation of pyrene labeled polyacrylic acid.The confonnation of PyP AA in alumina supernatant isshown in Fig. 7 as a function of pH. It can be seen that thepolymer chain becomes stretched out with increase in pHdue to the increasing intramolecular electrostatic repulsioncaused by the ionization of -COOH groups. The change inconformation of P AA upon the addition of the cationic poly-mer is also shown in Fig. 7. The coiling index shows anincrease followed by a decrease with the addition of the

cationic polymer. Coiling of polyacrylic acid with PDA-DMAC at low dosage is attributed to the charge neutraliza-tion of the acrylate groups by the oppositely charged cationicpolymer. With further addition of the cationic polymer. thepolymer complex could become positively charged with re-sultant restretching.

The change of PAA conformation with the addition ofPDADMAC clearly shows complexation between these twooppositely charged polymers. Such complexation is also con-firmed by the observation that higher adsorption density ofpolyacrylic acid is obtained on alumina with PDADMACpreadsorption than that on bare alumina at any pH condi-tions.

~. ofPDADMAC, AIIn0.1 I 10 100 1-

0.55 . '" . .. ..", ... ..,.., ",.

Correlation of Flocculation Responsewith Polymer Conformation

Results obtained for conformation of PAA at the solid-liquid interface under corresponding conditions are givenin Fig. 8. Interestingly, the effect of PDADMAC on theconformation of P AA is very different from that in solution.The coiling index of P AA is found to continuously decreasewith the addition of PDADMAC. In the case of flocculationwith polyacrylic acid alone, good flocculation could not beachieved because at low pH adsorption of the polymer inthe coiled form is unable to cause good bridging and at highpH the stretched polymer is unable to interact strongly withthe solid surface due to the electrostatic repulsion. In floccu-lation with dual polymers, unlike that with PAA alone, alu-mina particles are subjected to charge neutralization andreversal with PDADMAC preadsorption. While the zeta po-tential of alumina changes from - 30 to +40 m V due tothe adsorption of the cationic polymer, the PAA moleculesremain in the same stretched conformation since the solution

0.5

O.4S

0.4

0.3S

0.3

0.2S

0e~~~J'0u

"0

0.2

O.IS I . . . , . I I . . I

II 10 9 8 7 6 S

pH

FIG. 7. Confomlation change of PAA in solution as a function of pH(~) and concentration of PDADMAC (0).

POL YMER CONFORMA nON IN PLOCCULA nON 287

pH remains unchanged. The negatively charged stretchedP AA adsorbs on particles via interaction with preadsobedPDADMAC on alumina surface and provides better interpar-ticle bridging and thus excellent flocculation.

interparticle bridging when adsorbed on particles with pread-sorbed polymer of cationic charge.

ACKNOWLEDGMENT

The financial supports from die National Science Foundation and NalcoChemical Co. are deeply acknowledged.CONCLUSIONS

REFERENC~1. Confonnation of the adsorbed polymer molecules atthe solid-liquid interface can be altered drastically by theaddition of a secondary polymer as well as other changes insolution conditions such as pH.

2. Polymer-polymer complexation in the solution and atthe interface can lead to coadsorption of one polymer whichotherwise does not adsorb.

3. Use of dual polymer systems can markedly enhanceflocculation at a much lower dosage.

4. In addition to adsorption density, polymer confonna-tion plays a predominant role in flocculation. In the casestudied here, the stretched anionic polymer provides better

I. Montgomery, J~ "Precipitation, Coagulation and Flocculation inWaste Treatment principle and Design." Wiley, New York, 1985.

2. Britt, K. W., Dillon, A. G., and Evans, L. A., Tappi J. 60(7), 102( 1977).

3. Shimabayashi, S., Nishino, K., and Nakagaki, M., CoUoids Surf 63,121 (1992).

4. Somasundaran, P., Tjipangandjara, K. F., and Maltesh, C., in "SolidILiquid Separation: Waste Management and PrOOuctivity Enhance-ment" (H. S. Muralidhara, Ed.), p. 325. Battelle Press, Columbus,Ohio, 1989.

5. Tjipangandjara, K. F., and Somasundaran, P., Colloids Surf 55, 245(1991).

6. Yu, X., and Somasundaran, P., Colloids Surf al. 17 (1993).