Exchange of Cationic Polyacrylamides Adsorbed on Monodisperse Polystyrene Latex and Cellulose Fibers: Effect of Molecular Weight HIROO TANAKA, 1 LARS ODBERG, 2 LARS WAGBERG, 3 AND T O M L I N D S T R O M 4 Swedish Pulp and Paper Research Institute, Box 5604, S-I14 86 Stockholm, Sweden
Received January 23, 1989; accepted April 5, 1989 The exchange reactions of cationic polyacrylamides(C-PAM; 1.4 meq/g) on monodisperse polystyrene latex (PSL) and cellulose fibers have been studied using fluorescently labeled polymers. For C-PAM with the highest molecular mass (8 X 106) no significant exchange reaction could be detected on PSL or on fibers. When C-PAM with medium molecular mass (4 X 10s) was used, a moderate exchange reaction occurred. A significant exchange reaction took place, however, when C-PAM with a relatively low molecular mass (2 X 104) was used. The exchange reactions initially proceeded quite rapidly and then slowed down. The initial exchange was more extensive on fibers than on PSL, probably because of the lower surface charge density of the fibers, but it leveled off rapidly. This difference is most probably an effect of the porous structure of the cellulose fibers. © 1990AcademicPress,Inc. INTRODUCTION Studies o f p o l y m e r exchange reactions on solid surfaces should provide insight into the processes occurring during adsorption and also give a better understanding o f the polymerinduced stability or instability o f colloidal systems. These studies might also indicate whether p o l y m e r adsorption is a reversible process. It has been indicated that the p o l y m e r exchange reactions take place when the segment-surface binding energies o f displacing polymers are higher than those o f adsorbed ones ( 1 - 2 ) . There have also been several reports ( 3 - 9 ) o f the preferential adsorption o f high molecular mass species of n o n i o n i c polymers. Recently Pefferkorn et al. reported that a small fraction o f radioactively labeled polyacrylamide loosely 1Permanent address: Faculty of Agriculture, Kyushu University, Higashi-ku, Fukuoka 812, Japan. 2 To whom all correspondence should be addressed. 3 Present address: SCA Teknik AB, Box 3054, S-850 03 Sundsvall, Sweden. 4 Present address: MoDo AB, S-891 91 Ornsk61dsvik, Sweden.
attached to n o n p o r o u s glass beads exchanged very slowly with unlabeled molecules (10). Studies on the preferential adsorption or displacement reactions of polyelectrolytes are, however, very limited in number. Bain et al. ( 1 1 ) reported that in the case o f polydisperse anionic polyelectrolytes such as sodium polyacrylate and sodium carboxymethyl cellulose on BaSO4, low molecular mass polymers were preferentially adsorbed and were not displaced by high molecular mass p o l y m e r even after 71 days. Similar results were obtained for CaCO3 by A d a m and R o b b (12). Recently Wright et al. (13) studied the desorption o f polyelectrolytes from b a r i u m sulphate in the presence o f various displacing species and reported that, when polyacrylate was both the adsorbed and displacing species, no significant desorption occurred when the dispersion m e d i u m was water. Significant desorption occurred in 0.1 mole d m 3 NaC1, but only when the displacing polyacrylate had a higher molecular mass than that adsorbed. Studies have so far been concerned with nonionic and anionic polymers. The behavior o f cationic polymers on anionic surfaces are par-
229 0021-9797/90 $3.00 Journal of Colloid and Interface Science, Vol. 134, No. 1, January 1990
Copyright © 1990 by Academic Press, Inc. All rights of reproduction in any form reserved.
T A N A K A ET AL.
ticularly interesting from a practical point o f view since most natural surfaces have an anionic character. The purpose of the present investigation has been to study the exchange reactions of CPAM on monodisperse polystyrene latex and cellulose fibers. This was done using fluorescently labeled and unlabeled polymers. The synthesis of these polymers over a wide range of molecular weights and charge densities has been reported earlier (14). We have also reported the adsorption of these polymers on PSL and cellulose fibers ( 15 ).
The cellulose fibers used were from a neverdried, bleached, softwood kraft pulp, Imperial Anchor (Iggesund, Sweden). The fine material was removed before use. The charge density of the pulp as measured by conductometric titration (18) was 29 t~eq/g. Methods and Procedures
All experiments were performed at room temperature (22 + I°C) with gentle stirring using a magnetic stirrer. Adsorption was carried out so as to allow about 40% of the added DC-PAM to be adsorbed on PSL or cellulose fibers on the basis EXPERIMENTAL of the data reported earlier (15). Reaction Materials conditions and adsorbed amounts are given C-PAM samples with molecular masses of in Table II and III. 2 × 10 4, 4 × 105, or 8 × 10 6, estimated from The contact time between DC-PAM and intrinsic viscosity (16), were prepared and PSL was 2-3 h. In the case of fibers, times fluorescently labeled by dansylation as de- from 1 day to 2 weeks were used for equiliscribed previously ( 14, 15 ). The characteristics bration, depending on the molecular mass of of C-PAM and the dansylated C-PAM (DC- polymer. After adsorption, the PSL, except for PAM) are listed in Table I. The degree of la- the PSL-DC-PAM (8 × 106) system, was filbeling was less than 1 mole% and did not affect tered using a membrane with an average pore the adsorption behavior on either PSL or cel- diameter of 0.6/~m (Schleicher & Schfill, diam lulose fibers. = 25 m m ) and washed with deionized water PSL with a diameter of 925 nm and charge until no fluorescence was detected in the fildensity of 3.5 tzeq/g (5.46 # C / c m 2) was syn- trate. In the case of the PSL-DC-PAM (8 thesized according to (17). Details are given X 106) system, the PSL with adsorbed polymer in the previous report ( 15 ). was centrifuged at 20,000 rpm for 30 min in TABLE I Properties of Cationic Polyacrylamides Molecularmass Code P1 P*I P3 P*3 P5 P*5
N Y N Y N Y
0.19 0.17 1.55 1.59 12.5 13.6
b 1.7 1.4 3.2 3.4 6.1 6.7
X X X X X X
c 104 l04 105 105 106 106
2.9 2.5 5.2 5.3 9.0 10.0
X X X X X X
104 104 105 105 106 l06
1.43 1.41 1.38 1.42 1.40 1.49
Note. The molecular masses of P1 (P*I), P3 (P*3), P5 (P*5) are in the text given as 2 X 104, 4 X 105, and 8 X 106, respectively. a N, unlabeled; Y, labeled. b From [n]~/~c N,a i M = 1.91 X 10-4 Mww0"71 for PAM (Ref. (16)). c From [,1]~/~c Naa 1 M = 1.05 X 10 4 ~ww0.73 for (AM-CMA) 30% (Ref. (16)). Journal of ColloM and Interface Science, Vol. 134,No. i, January 1990
EXCHANGE OF CATIONIC POLYACRYLAMIDES TABLE II Conditions and Results for Adsorption of DC-PAM and Conditions for Polymer Exchange on PSL
Initial conc. of DC-PAM (nag/l)
Conc. of DC-PAM at equilibrium (nag/liter)
DC-PAM adsorbed on PSL (mg/g)
Conc. of C-PAM as displacer (mg/liter)
C-PAM DC-PAM (charge ratio)
P*I P*3 P*5
87 86 85
51 49 52
3.0 3.1 2.7
355 380 345
10 10 10
Note. P S L 1.20%, 22 + I ° C , p H 4.5 ~ 5.0.
a Beckman L5-40, decanted, and soaked in water. This procedure was repeated until no fluorescence was detected in the supernatant. The aqueous suspension of the resulting PSL had a solids content of 1.44%. The cellulosic fibers treated by DC-PAM were filtrated using a membrane with 3 ~m pore diameter (Nuclepore, diam = 25 m m ) and washed extensively. The aqueous suspensions of the cellulosic fibers had a solids content of 0.6%. The amounts of adsorbed polymer were determined by colloid titration (19) and from the residual fluorescence intensity of the filtrate or the supernatant. In the exchange experiments, 10 ml of aqueous solution of unlabeled C-PAM with the same molecular mass and charge density as the adsorbed DC-PAM was added to 50 ml of PSL or cellulosic fiber suspension. The molar ratio of C-PAM to DC-PAM was 10:1. The suspension was stirred and, at proper time intervals, 5-6 ml of suspension was withdrawn and filtered or centrifuged for complete sep-
aration. The fluorescence intensity in this filtrate or supernatant was measured using a Perkin-Elmer LS-5 Luminescence spectrometer. The excitation and emission wavelengths used were 333 and 538 nm, respectively, as described previously (14). The intensities were compared with the calibration curves to determine the desorbed amount of DC-PAM. RESULTS
Exchange Reaction of C-PAM on PSL Figure 1 shows the results of the polymer exchange reactions for combinations of the same molecular species except that the preadsorbed polymers were fluorescently labeled. CPAMs with three different molecular masses (2 X 104, 4 X 105, and 8 X 106) but with very similar charge densities were used. The results of blank tests without displacing polymers are also given in Fig. 1, in which it can be seen that practically no desorption can be detected when no displacer is present.
T A B L E IIl C o n d i t i o n s a n d R e s u l t s f o r A d s o r p t i o n o f D C - P A M a n d C o n d i t i o n s for P o l y m e r E x c h a n g e o n Cellulose F i b e r
Initial conc. of DC-PAM (mg/liter)
Conc, of DC-PAM at equifibrium (mg/liter)
DC-PAM adsorbed on fiber (rag/g)
Conc. of C-PAM as displacer (nag/liter)
C-PAM DC-PAM (charge ratio)
P*I P*3 P*5
335 190 75
187 109 45
29.6 16.2 6.0
1460 834 319
10 10 10
Note. Cellulose fiber 5 g/liter, 22 + I ° C , p H 4.5 ~ 5.0. Journal of Colloid and Interface Science, Vol. 134, No. 1, January 1990
T A N A K A ET AL.
60 iI sI
~¢- 20 .2
10 - vz20 Time, days
FIG. 1. The fraction of desorbed fluorescently labeled polyelectrolyte, DC-PAM, on PSL for various molecular weights. Reaction conditions see Table II. (©) P * I - P 1 , (A) P*3-P3, (lq) P*5-P5. Solid symbols for blank tests with no addition of displacer. Details of polymers see Table I.
In the case of the polymer with the highest molecular mass (P*5) no desorption can be detected even when displacer polymer is present. This lack of exchange agrees with previous investigations (12) and may be explained by the following simple considerations. First, the fact that the number of charges per molecule for P*5 is of the order of 1 × 10 4 m e a n s that every chain may have this many linkages to the surface. It is highly unlikely that all these linkages will be broken at the same time, especially if no polymer that can compete for these surface sites is present in solution. Second, the interaction between the charged segments of the polymer and the PSL-surface is electrostatic in nature and thus gives a very strong binding. In the exchange reactions with medium molecular mass (P*3) and low molecular mass ( P * I ) polymer, a considerable desorption can be detected. For P*3, 9% may be desorbed over a period of 40 days and for P*I, 60% may be des0rbed after 32 days. For both polyJournal of Colloid and Interface Science, Vol. 134,No. 1, January 1990
mers there is a high initial desorption but the process continues even after 40 days. The numbers of charges for P*3 and P*I and hence the numbers of possible linkages per molecule are 570 and 20, respectively. These numbers can explain why these polymers are exchanged when P*5 is not and they may also explain the difference in exchange between P*3 and P*I. Several factors may contribute to the shape of the exchange curves. There is probably a distribution of conformations of the polymers on the surface and some polymers with a higher number of interacting segments may be firmly anchored to the surface. The probability of desorbing these polymers will be low and a long time will be needed before they are exchanged with polymers in solution. These firmly anchored polymers are probably those that were first attached to the surface. A second factor, though less important in the present case where the excess of unlabeled polymer is very large, is that desorbed fluorescently labeled polymers may start to exchange with unlabeled polymers on the surface. This will result in a slower apparent exchange rate when the concentration of desorbed labeled polymers becomes high. It is not believed that surface roughness will contribute to the slow release over such a long period of time. Since the polymer molecules have estimated sizes of 50 and 15 nm, respectively ( 15 ), the PSL particles, with diameters of 925 nm, should appear as large, smooth surfaces to the approaching molecules. The results obtained suggest that although the polyelectrolyte interacts with many charged groups on the surface the situation is dynamic and the polymer is constantly rearranging itself. One or more of the binding sites may start interacting strongly with an approaching molecule in solution and this approaching molecule may gradually occupy more binding sites and eventually displace the originally adsorbed molecule. This process is, however, extremely slow for high molecular mass polymers.
EXCHANGE OF CATIONIC POLYACRYLAMIDES E x c h a n g e R e a c t i o n o f C - P A M on Cellulose Fibers
The results of exchange reactions of CPAMs on cellulose fibers are given in Fig. 2. Without displacing polymer small amounts of polymers P * I and P*3 were desorbed. No desorption occurred in the case of the highest molecular weight polymer, P*5. A slight exchange reaction, less than 1% exchange, of P*5 occurred in the presence of P5, which suggests that some weakly adsorbed P*5 polymers are present on the fibers. The exchange of polymer with m e d i u m molecular mass (4 × 105) also took place more easily on cellulose fibers than on PSL. As seen in Fig. 2, the degree of exchange on fiber reached 25% after 10 days compared to 7% on PSL. However, the degree of exchange leveled offclose to 25%. The exchange of polymer of low molecular mass (P* 1 ) initially occurred quite easily but leveled offclose to 47% after 2 days, as shown in Fig. 2. This situation differs considerably from that of P * I on PSL, shown in Fig. 1. The difference most probably originates in the dif-
50 S -6 .~ aO
ferent surface structures, the porous structure of cellulose fibers, and the smooth surface of PSL. In our previous work ( 15 ), it was found that the adsorption equilibrium of C-PAMs onto PSL was achieved almost instantaneously after mixing, whereas equilibrium on cellulosic fibers was reached after a few hours up to several days depending on the molecular mass. This indicates that the penetration of polymers into the pores is difficult and once inside the pores the polymers are also difficult to exchange. On the other hand, the polymers adsorbed on outer surfaces of cellulosic fibers are easier to exchange than those on PSL because of a weaker interaction, as estimated from the charge densities of cellulose fibers (1.4 /~C/ cm 2) and PSL ( 5 . 4 6 / z C / c m 2 ) . CONCLUSIONS The results show that the exchange of polymers on surfaces depends not only on the molecular mass of the adsorbate but also on the charge densities and solution concentration of the polymers in both the adsorption and exchange situations. The surface characteristics are, of course, also of great importance. The fact that an exchange does indeed occur indicates that the polymers adsorbed are in a state of dynamic equilibrium. The effect on the exchange reaction of other displacers and promoters such as added salt is also of great interest and such studies are currently in progress. REFERENCES
~u lo, U0 _t~-
10 15 Time, days
FIG. 2. The fraction of desorbed fluorescentlylabeled polyelectrolyte, DC-PAM, on cellulosic fibers for various molecular weights. Reaction conditions see Table III. ((3) P*I-P1, (A) P*3-P3, ([~) P*5-P5. Solid symbols for blank tests with no addition of displacer. Details of polymers see Table I.
1. Roe, R. J., in "Adhesion and Adsorption of Polymers, Part B" (L. H. Lee, Ed.), p. 629. Plenum, New York/London, 1980. 2. Botham,R., and Thies, C., J. Polym Sci. Polym. Syrup. C30, 369 (1970). 3. Felter, R. E., Moyer, E. S., and Ray, L. N., J. Polym. Sci. B7, 529 (1969). 4. Felter, R. E., and Ray, L. N., J. Colloid Interface Sci. 32, 349 (1970). 5. Howard, G. J., and Woods, S. J., J. Polym. Sci. A10, 1023 (1972). 6. van der Linden, C., and van Leemput, R., J. Colloid Interface Sci. 67, 63 (1978). Journal of Colloid and Interface Science,
Vol. 134,No. 1,January 1990
TANAKA ET AL.
7. Furusawa, K., and Yamamoto, K., J. Colloid Interface Sci. 96, 268 (1983). 8. Furusawa, K., and Yamamoto, K., Bull. Chem. Soc. Japan 56, 1958 (1983). 9. Kawaguchi, M., and Takahashi, A., Macromolecules 17, 1666 (1984). 10. Pefferkorn, E., Carroy, A., and Varoqui, R., J. Polym. Sci., Part B: Polym. Phys. 23, 1997 (1985). 11. Bain, D. R., Cafe, M. C., Robb, I. D., and Williams, P. A., J. Colloid Interface Sci., 88, 467 (1982). 12. Adam, V. S., and Robb, I. D., J. Chem. Soc. Faraday Trans. 1 79, 2745 (1983). 13. Wright, J. A., Harrop, R., and Williams, P. A., Colloids Surf. 24, 249 (1987). 14. Tanaka, H., and Odberg, L., J. Polymer Sci, Polymer Chem.
Journal of Colloid and Interface Science, Vol. 134, No. 1, January 1990
15. Tanaka, H., ()dberg, L., Whgberg, L., and Lindstr6m, T., J. Colloid Interface Sci. 134, 219 (1990). 16. Mabire, F., Audebert, R., and Quivoron, C., Polymer 25, 1317 (1984). 17. Cosgrove, T., private communication. 18. Katz, S., Beatson, R. T., and Scallan, A. M., Sven. Papperstidn. 87, R48 (1984). 19. Senju, R., Bull Chem. Soc. Japan 26, 143 (1953); Senju, R., "Koroido Tekitei Ho." Nankodo, Tokyo, 1969. [in Japanese] 20. Michaels, A. S., Mir, L., and Schneider, N. S., J. Phys. Chem. 69, 1447 ( 1965); 69, 1456 ( 1965); Tsuchida, E., Osada, Y., and Sanada, K., J. Polym. Sci. 10, 3397 (1972).