2108 Reprinted from LANGMUIR, 1991, 7.Copyright @ 1991 by the American Chemical Society and reprinted by permission of the copyright owner.

Polymer-Polymer Complexation in Dilute AqueousSolutions: Poly(acrylic acid)-Poly(ethylene oxide) and

Poly(acrylic acid)-Poly(vinylpyrrolidone)

PradipTata Research Development & Design Center, 1 Mangaldas Road, Pune 411001, India

C. Maltesh and P. Somasundaran.

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

R. A. Kulkarni and S. Gundiah

Polymer Chemistry Division, National Chemical Laboratory, Pune 411008, India

Received February 11,1991. In Final Form: May 22,1991

Fluorescence spectroscopy of pyrene-labeled poly(acrylic acid) (P AA) was used to study its interactionswith poly (ethylene oxide) (PEO) and poly(vinylpyrrolidone) (PVP) in aqueous solutions. The effects ofmolar ratio. molecular weight, and solvent were also investigated. Hydrogen bonding is the primarymechanism of interaction among these polymers. Excimer fluorescence studies show that the PVP-P AAcomplex is stronger than and exists in a more constricted form than the PEo-P AA complex. Interactionsbetween PVP and PAA are prevalent over a wider pH range than those of PAA and PEa. The higherelectronegativity of the oxygen in the pyrrolidone group is attributed as the reason for the stronger interactionof PVP with poly(acrylic acid).

Introduction

Cooperative interaction between synthetic water-solublepolymers has been a topic of investigation worldwidebecause of their importance from both scientific andpractical viewpoints. Interpolymer complexes resultingfrom these interactions possess unique properties whichare different from those of the individual components andfind various applications, for example, in dialysis, ultrafil-tration, reverse osmosis, and production of blood-compatible materials and batteries.1,2 Synthetic water-soluble polymers such as polyacrylamide, poly(acrylicacid), poly( ethylene oxide), and poly( vinylpyrrolidone) arealso increasingly being used in mineral-processing oper-ations such as flocculation and dewatering, effluenttreatment, selective flocculation,3 and flotation.4 A naturaldevelopment in this area has been the use of polymermixtures in mineral processing. In a recent study, poly-(vinylpyrrolidone) (PVP) was found to be a better dis-persant than poly (ethylene oxide) (PEO) and sodiumsilicate for selective flocculation of iron oxide from clayusing poly (acrylic acid) (PAA) as flocculant.5 Complex-ation between PVP, PEa, and P AA has been investigatedearlier but the studies have concentrated on the biologicalaspect of these complexes. The interactive forces respon-sible for complexation between two chemically differentmacromolecules are usually secondary forces due toelectrostatic, hydrogen-bonding, or hydrophobic interac-tions. Hydrogen bonding in the poly(acrylic acid)-poly-

(ethylene oxide) system is perhaps the most well estab-lished6,7 and is used as a quantitative method for estimatingppm levels of PEO or P AA. Earlier studies suggested theassociation to be between the ether oxygen of the poly-(ethylene oxide) and the carboxyl group of the poly(acrylicacid), which approaches 1:lstoichiometry.8 Cooperativeinteraction among active sites in PEQ-P AA complexformation was later verified by several techniques.9 Ithas also been suggested that the driving force for theformation of a hydrogen bond between an ether and acarboxylic acid in aqueous solution is very small and astable PAA-PEO complex can be formed only by thecooperative interactions between many such groups. 10 Theformation of hydrogen bonds is also considered to be thedominant force in the interactions between PVP andP AA.U.12 A stronger association of P AA-PVP than thatof PEQ-P AA was observed and attributed to the contri-bution of hydrophobic and Coulombic interactions in thePAA-PVP association.

The present study was conducted to investigate theinteractions between P AA and PEO and between P AAand PVP using fluorescence spectroscopy. Effects ofchanges in complexation conditions such as pH, polymermolecular weight, and solvent are also discussed. It hasbeen demonstrated that fluorescence spectroscopy is amore sensitive and informative technique than otherconventional methods such as viscometry, potentiome-try, and turbidimetry, which provide information only onaverage properties representing the whole system.13-18

(6) Smith, K. L.; Winslow, A. E.; Petel8en. D. E.1 rid. E",. Chem. 195t,51, 1361.

(7) Ikawa, T.; Abe, K.; Honda, K.; Tsuchida, E. J. Polym. &i., Polym.Chem. Ed. 1175, 13, 1505-1514.

(8) Bailey, F. E., Jr.; Lundberg, R. D.; Callard, R. W. J. Polym. Sci.Part A 1964,2,845-851.

(9) Ika-. T.; Abe, K.; Honda, K.; Tauchida, E. J. Polym. Sci., Polym.Chem. Ed. 1975, 13, 1505-1514.

(10) Oyama, H. T.; Tang, W. T.; Frank, C. W. Macromoleculellt87,20,474-480; 1839-1847.

(11) 0Iada. Y. J. Polym. Sci., Polym. Chem. Ed. 1179,17,3485.(12) Pierola, I. F.; Caceres, M.; Caceres, P.; CasteUanOB. M. A.; Nunez,

J. Eur. Poiym. J. It88, 9, 895-898.

(1) Bekturov, E. A., Bimendina, L. A. Adu. Polym. Sa. ItSl, 41, 9&-1.7.

(2) Tauchida, E.; Abe, K. Adu. Polym. Sci. 1982,45,1-119.(3) Somasundaran, P.; Wang, Y. H. C.; Acar, S. In Future Trends

in Polymer Science and Technology. Polymers: Commodities orSpecialties?; Martuscelli, E., Marchetta, C., Nicolais, L., Eda.; Tech-nomic Publishing Inc.: Berlin, 1987, pp 134-155.

(.) Colombo, A. F. In Fine Particles Processing; Somaaundaran, P.,Ed.; AIME: New YOI'k. 1980; pp 1034-1056.

(5) Raviahankar, S. A.; Pradip. Role of Organic Diperaanta in theSelective Flocculation of Iron Oxide/Kaolinite Mixture&. Preaented atthe 62nd Colloid and Surface Science Symposium, American Chemic:alSociety, June 1988.

0743-7463/91/2407-2108$02.50/0 @ 1991 American Chemical Society

Langmuir, Vol. 7, No. 10, 1991 2109Polymer-Polymer Complexation

Table I. CbaJ'aeteristj~ of the Polymen Ueed ia the Study

MWXI0""4 notation pyrene content, wt % supplierpolymerpyrene-labeled poly(acrylic acid)poiy(ethylene oxide)poly(ethylene oxide)poly( vinYlpyrrolidone)poly(vinyl pyrrolidone)

1.35 laboratory synthesisPolymer Lab. Ltd., U.K.Scientific Polymer Products, Inc., New YorkSigma Chemicals Co., St. Louis, MOAldrich Chemical Co., Milwaukee, WI

2 3 4 5 6 7 8 9 10 11 12

pH

Figure 1. Interactions between poly(acrylic acid) and poiy-(ethylene oxide) of two molecular weights, 100 000 and 865 000.

Experimental Section

Synthesis of 2-[4-(1-pyrenyl)butanoyl]aminopropenoic acid,which was used as the pyrene source, is described elsewhere.!1Acrylic acid monomer purchased from Aldrich was vacuumdistilled at 50-51 °C to free it from inhibitor. It was thencopolymerized with pyrene-containing monomer (2-[4-(1-pyre-nyl)butanoyl]aminopropenoic acid) in DMF at 65 °C withazobis-(isobutyronitrile) as the initiator for 9 h. The concentrations ofthe acrylic acid and the pyrene-containing monomer were 72.8and 1.07 mM, respectively. The concentration of AIBN was 0.3mol %,andtheamountofDMFwas35.5mL. Thepyrenecontentof the polymers was estimated from UV absorption Spectra.!1Intrinsic viscosities were obtained from linear !Iop/ C VB C plotsand used to determine molecular weights by use of Mark-Hou-wink constants available in the literature.18

Poly( ethylene oxide) and poly( vinylpyrrolidone) were obtainedfrom various suppliers and used as received. Characteristics ofthe polymers used are presented in Table L

The concentration of polymers in all the fluorescence exper-iments was maintained at 10 ppm. Only in cases where the effectof molar ratio was studied, was the concentration varied.

All samples were prepared in triply distilled water and at asalt concentration of 0.005 M sodium chloride.

Fisher Scientific ACS reagent grade hydrochloric acid andsodium hydroxide were used for pH adjustments. Urea crystalswere purchased from Amend Drug and Chemical Co., and dim-ethylsulfoxide (DMSO) was from Fisher Scientific. Both wereACS reagent grade.

Fluorescence spectra were recorded on a Photon Technology~100spectrophotometer. The excitation wavelength was 335Dm. Monomer emission was monitored at 376 Dm and excimeremission at 480 Dm. The sample cells were of 10-mm path length,and suitable correction was applied to avoid interference fromthe lamp.

Results and DiscussionPoly(acrylic acid) (P AA) behaves as a random coil in

dilute solutions in the presence of simple electrolytes. Therandomly distributed pendant pyrene groups on thepolymer chain, at low concentrations, do not significantlyaffect the dissolution characteristics. Pyrene groups areconstantly in motion due to the high mobility of thepolymer segments to which they are covalently attached.The distance between the pyrene groups on the polymerchain is determined by its conformation and the mobilityof the segments. An excited pyrene group interacts witha ground-state pyrene group to form an excimer whenthey approach each other within about 4-5 A. Theformation of an excimer will depend upon the conformationof the polymer, and hence, the extent of intramolecularexcimer formation (given by the ratio of excimer tomonomer intensities or 1./ I~ provides a measure of thestatistical conformation of the labeled polymer chain andits intramolecular overlap. It can therefore be referred toas the coiling index. A large value of the I./lm ratio

suggests polymer chain contraction, whereas a small valueof Ie/1m suggests polymer expansion and/or segmentalrigidity. Under the dilute polymer concentrations em-ployed, all excimer formation was intramolecular and hencethe Ie/1m value can be taken as a true indication of polymersegmental mobility.

The change in the conformation ofpoly(acrylic acid) ~a function of pH is shown in Figure 1. At low pH values,the acid groups of the P AA are undissociated and thereforethe polymer exists as a compact coil (high Ie/1m). Whenthe pH of the solution reaches the pK. of PAA (4-4.5),some of the COO H groups of the P AA are ionized. Thisgenerates some repulsion between adjacent groups and asa result the polymer coil expands. The presence of aneutral electrolyte shields some of the charge and decreasesthe repulsion. This is reported as a gradual decrease inthe Ie/1m value with increase in pH. Above a pH value of8.5 all the acid groups on the P AA are ionized and thepolymer is in its most expanded state. As a result, thereis no further decrease in I e/ 1m and it remains at a constantvalue.

It can also be see~ from Figure 1 that the conformationof poly(acrylic acid) is affected by the presence of poly-(ethylene oxide) (PEO), but only below pH 4.5. Undis-sociated COOH groups playa significant role in PAA-PEO interactions, and the presence of a few dissociatedacid groups on the P AA is sufficient to break up hydrogenbonding between P AA and PEO. The strength of a singlehydrogen bond between an ether oxygen of the poly-(ethylene oxide) and a carboxylic acid of the poly(acrylicacid) is very low. The total strength of a PAA-PEOcomplex will therefore depend upon the number ofhydrogen bonds formed as well as the distance betweentwo hydrogen-bonded groups. The interaction betweenP AA and PEO is a cooperative phenomenon, and thestability of the complex will depend on the predominanceof hydrogen bonding over thermal or Brownian motion.At low pH values «3) the number of dissociated COOHgroups on the P AA is very low and there are a large number

(13) Turro, N. J.; Arora, K. S. Polymer 19M, 27, 783-796.(14) Arora, K. S.; Turro, N. J.J. Polym. Sci., Part B 1981,25, 243-262.(15) Wanc, Y.; Morawetz, H. Macromolecule. It89,22, 164-167.(16) Heyward, J. J.; Ghi(ljno, K. P. Macromolecule.lt89, 22,11~

1165.(17) Oyama, H. T.; Hemker, D. J.; Frank, C. W. Macromolecule. 1989,

22, 1255-1260.(18) Hemker, D. J.; Garza, V.; Frank, C. W. Macromolecule. It90,23,

«11~18.(19) Kulbrni, R. A.; Gundiah, S. Makromol. Chem. It84, 185,957.

Pradip et at.2110 Langmuir, Vol. 7, No. 10, 1991

PVP molecular weight (Figure 2). In the complexationbetween P AA and PVP and between P AA and PEa, it isevident that the stability of the complex is determined bythe strength of the individual H bonds as well as thedistance between hydrogen- bonded groups. The strongerH bond in the PAA-PVP system results in that complexbeing stable over a wider pH range. Only above a pHvalue of 8, where all the acid groups of the P AA are ionized,are the interactions between P AA and PVP absent.Possible interactions between P AA, PEa, and PVP areschematically illustrated in Figure 3.

The effect of molar ratio of the interacting polymers(PAA-PEO and PAA-PVP) was also studied, and theresults are given in Figure 4. It has been stated that theinterpolymer complexes formed are stoichiometric. Thisis true for the P AA-PEO system, where it can be seen thatthe plateau Ie/1m value of 0.08 is reached at a molar ratioof 1.0. In the PAA-PVP system, however, completestretching of the complex occurs at a molar ratio as lowas 0.01. This result is very surprising, and at the presentmoment we do not have any explanation for this behavior.These results lend support to the earlier hypotheses thatthe interactions between PAA-PVP are stronger thanthose in the FAA-PEa system and that a stable PAA-PEa complex requires a 1:1 stoichiometry.

It was also observed that the interpolymer complexesformed were totally reversible. Instead of a new samplebeing prepared at every pH, the same sample was takenthrough a cycle of pH 2-10 and back. The ability to form,break, and re-form became clear from this behavior.

The solvent can play an important role in determiningthe nature of polymer complexes formed by hydrogenbonding because it also can interact with the polymer viahydrogen bonds.2 Complexation in the presence of suchsolvents, therefore, can be regarded as a three-componentreaction involving a proton-donating polymer, a proton-accepting polymer, and a solvent. The effect of such asolvent, dimethyl sulfoxide (DMSO), on polymer-polymercomplexation is depicted in Figure 5. The poly(acrylicacid) is mostly undissociated at a pH value of 3.0, andDMSO being a better solvent than water causes the coilto expand. The value of the coiling index for P AA alonein water is higher than that in DMSO. In the case of theP AA-PEO system, a combined effect of both the solventmolecules and PEO on the stretching of the P AA (lowerIe/1m than that for the PAA-PEO complex in water) ispossible at low DMSO content. As the DMSO content israised, interactions between DMSO and PAA becomestronger since the amount of DMSO is much larger thanthat of the proton-accepting groups of the poly(ethyleneoxide). Also, the mobility of the DMSO molecules is higherthan the segmental mobility of the poly(ethylene oxide)chain. Eventually, at 40% DMSO all FAA-PEa bondsare replaced by P AA-DMSO bonds. In the case of P AA-PVP, the bonding is stronger than that between PAA-PEa and hence a higher amount of DMSO is required toreplace the interaction between poly(acrylic acid) and poly-(vinylpyrrolidone) than is required to replace FAA-PEainteractions. Interpolymer complexation was also studiedin the presence of urea, but the effect was not aspronounced as that in the case of dimethyl sulfoxide.

Figure 2. Interactions between poly(acrylic acid} and poly(vi-nylpyrrolidone) of two molecular weights, 360 000 and 40 000.

of hydrogen bonds. The large number of hydrogen bondsand the small separation between hydrogen-bonded groupswill result in a strong complex and also prevent anysegmental mobility. The associated chains, therefore, arevery stiff and this results in a very low value of the coilingindex. With an increase in the pH, some of the acid groupsof the P AA will dissociate and break up the hydrogenbonds with the ether oxygen of the PEa at these points.As a result, the distance between some hydrogen-bondedgroups will increase and the segmental mobility of theassociated chains will also increase slightly. The measuredcoiling index will be slightly higher, as is seen in Figure1. Finally, at a pH value of 4.5, the number of dissociatedacid groups on the P AA is very high, the segmental mobilitybecomes large enough to overcome the strength of thehydrogen bonds, and the complex between P AA and PEOis broken. This can be seen from Figure 1, where the coilingindex of P AA above pH 4.5 is similar in the presence aswell as the absence ofPEO. It can also be seen from Figure1 that there is no effect of molecular weight of poly( ethyleneoxide) on its interactions with poly(acrylic acid), at leastin the ranges of molecular weight, concentration, and timeinvestigated in this study. This implies that the distanceof separation between hydrogen-bonded groups ratherthan the overall number of bonds is the critical factorgoverning the stability of the PAA-PEO complex.

In the presence of poly(vinylpyrrolidone) (PVP), thecoiling index (1.11 m) ofP AA reaches a value of zero (Figure2). This suggests that thePAAchainassociated withPVPis stiffer than the one associated with PEa. The pH rangeover which the P AA-PVP complex is stable is wider thanthe corresponding range for the PAA-PEO complex.These results suggest that the interactions between P AAand PVP are stronger than those between P AA and PEa.The hydrogen bonding between P AA and PVP could bestronger than that between P AA and PEO due to the higherelectronegativity of the oxygen of the PVP caused by thepresence of nitrogen in the functional group. Thereforethe PAA-PVP complex will remain stable with fewerhydrogen bonds and over a wider pH range than the P AA-PEO compleL During interactions between P AA andPVP, the bulky functional group of the PVP could causesteric hindrances leading to a more rigid chain (resultingin a lower leI 1m). The PVP ring is known to be hydro-phobic (presence of three methylene groups), but this doesnot appear to playa significant role in its interactionswith P AA, as can be postulated from the lack of effect of

Conclusions

Interactions between poly(acrylic acid), poly(ethyleneoxide), and poly(vinylpyrrolidone) in aqueous solutionwere studied by fluorescence spectroscopy. Hydrogenbonding is the primary mechanism of interaction betweenthese polymers, and the interactions between P AA and

Langmuir, Vol. 7, No. 10, 1991 2111Polymer-Polymer Complexation

~ ~_~I~~I: -

~~~~~--I - I ~-

(yo (yoMany Bonds - No Mobility

--I - I ~-

N N

O't 0'1. I

--o~ ~~o-~IiI

(belCRI 3)

pHbe~3and 4.5 pHbe~3

and 6

No Bonding

~'~~o~

pH atX>ve 7

--YX-

--I~ 1-

(yo (yo-- o~ o~O" -

PVP-PAA

0.20PM ALONEPM + PEOPM + PVP

06cO

0.15_2 pH = 3.0 t 0.1

_w

'0~ .

~

XraJ 0.10Q!:

0z::3

'0(.) 0.05

PEO-PAAFipn 3. Schematic representation of interpolyJDer complexation.

0.22

0.20

0.18

..1' 0.1'".JI 0.14,t1 0.12

Q!: 0.10

(.)~ 0.08

0u 0.06

O.O~

0.02

0.000.0 0.5 1.0 1.5 2.0

MOLAR RATIO: (PEO or PVP) / PM

FiIw'e 4. Effect of molar ratio on interpolymer complexation.PVP were stronger than those between P AA and PEO.This was attributed to the stronger hydrogen-bondingcapability of the pyrrolidone group in the PAA-PVPsystem. In the presence of DMSO, the interpolymercomplexes were replaced by a solvent-P AA complex athigh solvent concentrations, the amount ofD MSO requiredbeing higher in the case of PVP than that in the case ofPEO. Formation of a stable P AA-PEO complex required1:1 stoichiometry but a stable PAA-PVP complex wasformed even at very low PVP to P AA ratios. The com-plexation process was found to be reversible upon chang-

~.oo.18eee'ee'e.a.L. I. I .,

0 10 20 30 40 SO 60 70 80 90 100

% DIASO

Figure S. Effect of solvent (DMSO) on interpolymer complex.ation.

iog pH of the solutions. Under the conditions studied,there was no detectable effect of molecular weight of eitherthe PRO or the PVP on its interaction with P AA.

Acknowledgment. Financial support from the Na-tional Science Foundation (Grant NSF-MSM-8617183)and the Department of Science & Technology (DST), NewDelhi, under the Indo-U.S. Science & Technology Ini-tiative Program is gratefully acknowledged.