- Email: [email protected]
Interactions between poly(ethylene oxide) and fatty acids sodium salts studied by surface tension measurements
Journal of Colloid and Interface Science 277 (2004) 215–220 www.elsevier.com/locate/jcis
Interactions between poly(ethylene oxide) and fatty acids sodium salts studied by surface tension measurements Elisa Zeno a,b,∗ , Davide Beneventi b , Bruno Carré b a Ecole Française de Papeterie et des Industries Graphiques (INPG), B.P. 65, 38402 St.-Martin d’Hères, France b Centre Technique du Papier, Ressources Fibreuses, Domaine Universitaire, B.P. 251, 38044 Grenoble Cedex 9, France
Received 20 January 2004; accepted 16 April 2004 Available online 19 May 2004
Abstract This work focuses onto the interactions between poly(ethylene oxide) (PEO) and fatty acids, in order to set their potential role of contaminants for PEO-based retention systems. Surface tension measurements were used to investigate PEO–fatty acid systems and they made it possible to clearly point out the interactions between the polymer and the sodium octadecylcarboxylates with different degrees of unsaturation. The observed interaction seems to be dependent on the fatty acids’ solubility, the increase of which leads to less pronounced phenomena, which are, in contrast, emphasized by the increase in PEO chain length. 2004 Elsevier Inc. All rights reserved. Keywords: Polyethylene oxide; Fatty acids; Surface tension; Degree of unsaturation
1. Introduction The papermaking process involves the use of different polymeric materials aimed to improve the fines and fillers retention in the fiber network. Retention aids can be either polyelectrolytes or nonionic polymers, such as the polyethylene oxide (PEO), which has been found to be an efficient retention aid if used in combination with an enhancer (usually a phenolic resin) [1–7]. During the past years, this system has been successfully applied with mechanical and recycled pulps, for which the conventional cationic retention systems failed because of the high levels of dissolved and colloidal substances [8–10]. Actually, cationic polymers interact with anionic substances originating from wood or carried over from the de-inking process, so that they are unavailable to develop their action onto the fibers. A nonionic retention aid such as PEO, being less sensitive to negative charges, should be able to maintain good retention performance even in the presence of high levels of anionic interfering substances. Nevertheless, several chemicals have been found to interfere with this system, in particular some of the nonionic * Corresponding author. Fax: +33-0-4-76-15-40-16.
E-mail address: [email protected] (E. Zeno). 0021-9797/$ – see front matter 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.jcis.2004.04.028
surfactants used in the de-inking lines and carried over to the paper machine [11]. Provided that the main residual de-inking chemicals carried over to the paper machine are the fatty acids [12–15] and that considerable amounts of these wood extractives can be dispersed or dissolved in the mechanical pulp after the mechanical fibrillation and bleaching [16], it seems important to investigate their interaction with the PEO, in order to set their potential role of contaminants for this retention system. Polymer–surfactant aqueous systems have been widely investigated over a long period with varied experimental techniques; most of these studies dealt with a neutral polymer and an anionic surfactant. In particular, the system of poly(ethylene oxide) (PEO) and sodium dodecyl sulfate (SDS) has been extensively investigated [17–34] and reviewed [35,36]. By means of classical physical and spectroscopic methods, the spontaneous association of the SDS and the PEO has been explained in terms of cooperative binding of the surfactant to the polymer in the form of aggregates, driven by hydrophobic and electrostatic interactions. According to the models proposed, the wrapping of the polymer around micelles would favor micellization, which starts at a concentration (critical aggregation concentration, c.a.c. or c1 ) lower than the critical micellar concentration (c.m.c.) and goes on up to the point of saturation of the polymer
216
E. Zeno et al. / Journal of Colloid and Interface Science 277 (2004) 215–220
(p.p.s. or c2 ). Moreover, the PEO capacity of binding the alkalimetal cations [37] would result in interaction with the counterion, leading to a positive charge distribution which interacts with the anionic surfactant. Referring to these models, some studies deal with the counterion effect on PEO–anionic surfactant complexation [32,33], whereas some others concern the effect of the alkyl chain length [19,20,38] and consequently of the surfactant hydrophobicity. More recently, the effect of the polar head group has been taken into account and the role of the carboxylate head group has been investigated [39], pointing out the stronger binding of the PEO with sodium dodecanoate. Furthermore, a recent study [38] reports the decrease in the PEO–sodium alkyl carboxylate interactions when the hydrophobicity of the surfactants is varied by decreasing the chain length from C12 to C8, where no interaction was observed. Actual results and the previous considerations of PEO use as a retention aid promoted this work, aimed at investigating the interaction between PEO and sodium octadecyl carboxylates with different degrees of unsaturation, and then focusing on the oleic acid, which is reported to be one of the dominant fatty acids in a mechanical pulp sample [16].
respectively the surfactant concentration, at a value above the c.m.c., and the polymer concentration, at 1 × 10−6 M. For the sodium oleate, some measurements were taken, keeping constant the molar ratio between PEO and NaOl. An UV/visible spectrophotometer was used to measure the light transmission of PEO–fatty acid systems. Measurements were performed at room temperature. The transmittance values at 260 nm were used to calculate a lightscattering coefficient according to Lambert’s law, I /I0 = e−γ x ,
(1)
where I is the intensity of the incident light, I0 the intensity of the incident light after passing through a thickness x of the liquid sample, and γ the extinction coefficient, which is composed of two terms, the scattering coefficient (τ ) and the absorption coefficient (κ), γ = τ + κ.
(2)
When light scattering is large compared to absorption, as in the case of PEO and fatty acid systems, the absorption coefficient can be neglected and Eq. (1) becomes I /I0 = e−τ x . The transmittance (T = I /I0 ) values are related to the scattering coefficient by the equation T = I /I0 = e−τ x ,
(3)
which was used to calculate τ .
2. Experimental 2.1. Materials and methods
3. Results and discussion High-purity commercial samples of poly(ethylene oxide) and poly(ethylene glycol) with average molecular weights of 194, 35 × 103, 1 × 105 , 6 × 105 , 9 × 105 , and 5 × 106 were used in this study. Sodium stearate (NaSt), sodium oleate (NaOl), and linolenic acid (NaLin) were all high purity grades (minimum 99%) and used as received. All solutions were prepared at pH 10 with concentrated commercial buffer solution (boric acid/potassium chloride/sodium hydroxide) and water purified with a Milli-Q unit. Linolenic acid dissolved at pH 10 in the presence of NaOH forms the corresponding sodium salt, so it is labeled (NaLin). Equilibrium surface tension measurements were performed with a De Noüy ring apparatus at 35 or 70 ◦ C. The adsorption isotherms of surfactants and polymers of different molecular weight were determined by measuring the surface tension value 3 min after the addition of increasing volumes of a concentrated mother solution to a fixed volume of deionized water. The ternary systems were investigated, keeping constant
3.1. Binary systems 3.1.1. Fatty acid sodium salts Adsorption isotherms showed that the increase in the fatty acid chain unsaturation led to an increase in the minimum surface tension and in the c.m.c. value (see Table 1). This behavior, in agreement with the findings of a previous work [40], can be interpreted as a result of the increase in hydrophilicity of the fatty acids, due to the chain unsaturation increase. 3.1.2. Poly(oxyethylene) Fig. 1 shows the results obtained for aqueous solutions of polyoxyethylene. For lower molecular-weights, the surface tension exhibits a strong molecular-weight dependence, while, for higher molecular weights, where the hydroxyl end group contributions become insignificant, all the samples
Table 1 Main properties of the fatty acids used in this study Fatty acid Stearic Oleic Linolenic
Formula CH3 –(CH2 )16 –CO2 H CH3 (CH2 )7 CH=CH(CH2 )7 CO2 H CH3 (CH2 CH=CH)3 (CH2 )7 CO2 H
Mp (◦ C) 69.6 13.4 −5
a As obtained from adsorption isotherms at pH 10 and 35 ◦ C (70 ◦ C for the stearic acid).
c.m.c.a (M) 7 × 10−5 7 × 10−5 7 × 10−4
γmin a (mN/m) 25.5 28 32.5
E. Zeno et al. / Journal of Colloid and Interface Science 277 (2004) 215–220
217
Table 2 Surface tension rise of the fatty acids micellar solution, subsequent to the addition of PEO Fatty acid
No. double bonds
γ − γmin (mN/m) 70 ◦ C 35 ◦ C
Stearic Oleic Linolenic
C (18:0) C (18:1) C (18:3)
26.9 11.1 –b
–a 20.5 4.2
a T ◦ Krafft = 66 C. b Difference within experimental error.
Fig. 1. Surface activity–molecular weight dependence of polyoxyethylene aqueous solutions (35 ◦ C).
Fig. 3. Schematic surface tension/log concentration plot of a surfactant (lower dotted line) and of a surfactant in the presence of a complexing polymer (full line: idealized behavior and dotted lines: experimental behavior) according to [36].
Fig. 2. Effect of the PEO (Mw = 9 × 105 ) addition on a micellar solution of NaSt at 70 ◦ C (F), NaOl at 70 ◦ C (P) and at 35 ◦ C (Q), and NaLin at 35 ◦ C (!).
have the same behavior, i.e., a region where the surface tension (60–61 mN/m) no longer decreases increasing the polymer concentration. In accord with several findings reported in the literature [41,42], the stabilization of the surface tension of high-Mw PEO solutions was associated with the saturation of the air/water interface by the polymer chains. 3.2. Ternary systems When PEO (9 ×105 Mw ) was added to a micellar solution of each fatty acid, the surface tension increased when the polymer concentration increased. The shape of the surface tension versus PEO concentration curve shown in Fig. 2 led to the assumption that on adding low amounts of polymer to the soap solution, a soluble complex was formed and, when a certain polymer/surfactant ratio was reached, phase separation occurred. The increase in surface tension was therefore correlated with the decrease in the surfactant concentration in the aqueous phase. It should be noted that measurements were performed at a surfactant concentration slightly above the c.m.c., in order to avoid dilution effects due to the addition of the polymer. Moreover, in the range of concentrations tested, the polymer surface tension was constant at the plateau value (60–61 mN/m). The PEO–surfactant binding
seems to be weaker when the number of double bonds increases, as shown in Table 2. This result is in line with the finding of a recent work [38], where it was shown that decreasing the chain length of the sodium alkylcarboxylates, the polymer–surfactant interactions becomes weaker. Actually, both the decrease in the number of carbon atoms and the increase in the number of double bonds result in a decrease in the surfactant hydrophobicity, which is an important parameter governing polymer–surfactant complexation. It should be noted that measurements were carried out at pH 10, but, as preliminary experiments seem to indicate, results might be applicable to lower alkaline pH, given that fatty acids are under the dissociated form for pH > 7; as an example, pK = 4.95 for the oleic acid. Provided that (i) PEO forms complexes with fatty acid sodium soaps, (ii) the strength of the interaction is weaker when the degree of unsaturation increases, i.e., with the linolenic acid, and (iii) in paper mill process water the stearic acid is totally precipitated because of temperature (below the Krafft point) and because of the presence of calcium ions, a more detailed investigation of the polymer–surfactant interaction was carried out only for the sodium oleate. A common method for studying the polymer–surfactant interaction is surface tension measurement as a function of surfactant concentration at a constant polymer concentration. The classical result [36], if complexation occurs, is schematically represented in Fig. 3. The first breaking point is the T1 or c.a.c., where the interaction starts and the surfactant, which was present only as monomers up to the c.a.c., begins to micellize onto the polymer in the bulk. In the dilute range, up to the c.a.c., the
218
E. Zeno et al. / Journal of Colloid and Interface Science 277 (2004) 215–220
Fig. 4. Surface tension of solutions containing a constant PEO concentration (1 × 10−6 M) and increasing NaOl concentrations.
surface tension is always below that of the surfactant solution itself, because of the polymer surface activity or of its enhancing effect on the surfactant adsorption in the case of large attractive interactions between the two components (see the poly(vynilpyrrolidone) (PVP)-SDS system [43]). At the second transition, T2 , the polymer is saturated and there is an accumulation of surfactant monomers, resulting in a surface tension decrease up to T2 , where ordinary aqueous surfactant micelles are formed. In such a model the surface tension behavior could be explained by changes in the bulk phase and the polymer/surfactant complex has no surface activity. In our conditions, PEO concentration fixed at 1 × 10−6 M, only the first break point was observed (Fig. 4) with a progressive shift of T1 toward low sodium oleate concentrations with increasing PEO Mw , i.e., T1 = 1.8 × 10−5 M for PEO 35,000 and T1 = 2 × 10−6 for PEO 900,000, which was associated with an increase in the surfactant–polymer interactions with the PEO Mw . Considering that the surfactant concentration at the T2 point is influenced by the polymer concentration [22], the reason could be a too high polymer amount to reach saturation in the concentration range tested. This hypothesis is in agreement with the dependence of the concentration interval between T1 and T2 on several parameters such as pH, T ◦ , counterions and polymer amount, reported in the literature [22,44]. In order to investigate the effect of the PEO amount, further measurements were carried out, keeping constant the PEO/NaOl molar ratio and varying the ratio between 0.0001 and 0.1. The results obtained are shown in Fig. 5 and it could be seen that for a surfactant concentration above the c.m.c. the saturation point would be reached only for the lowest ratio, which corresponds to a polymer concentration of 2 × 10−8 M. Moreover, by increasing the ratio PEO/NaOl the complexation phenomenon occurred for lower surfactant concentration, indicating that the polymer amount plays also a role on the beginning of the complexation. The complex formation was confirmed also by the light transmission measurements, according to Eq. (3). The scattering coefficient of NaOl–PEO mixtures was calculated and plotted versus surfactant concentration (Fig. 6). The NaOl concentration at the breaking point is the same as the concentration at the break-
Fig. 5. Effect of the PEO/NaOl molar ratio on the surface tension of NaOl (Q) in the absence of PEO and for PEO/NaOl mixtures (E, 1, !).
Fig. 6. Scattering coefficient as calculated from Eq. (3) for PEO/NaOl mixtures.
Fig. 7. Effect of PEO addition on a micellar solution of NaOl, varying the polymer chain length.
point in the surface tension versus concentration plot shown in Fig. 5. The interaction between NaOl and PEO having been verified, the effect of the PEO amount, we investigated the influence of the PEO chain length on the complexation, from the tetrameric species to a PEO of average molecular weight 5 × 106. Their addition to a NaOl micellar solution gave the results illustrated in Fig. 7. The addition of the oligomers did not induce any change in the surface tension, even when
E. Zeno et al. / Journal of Colloid and Interface Science 277 (2004) 215–220
219
binding of the surfactants to the polymer could be at the base of the impairing effect of the fatty acids on the retention system constituted by PEO and phenol formaldehyde resin, observed in preliminary work.
Acknowledgment Fig. 8. Molar ratio PEO/NaOl at which is observed the rise in the surface tension of a micellar solution of NaOl when increasing amounts of PEO are added.
the oligomer–surfactant ratio was hugely increased (data not reported in the figure). At higher molecular weights, the surface tension rose and the polymer concentration at which the increase started was dependent on the PEO chain length. In Fig. 8 are reported the molar ratios between PEO and NaOl at which the surface tension increase began, as a function of the PEO chain length. It could be assumed that the chain length plays the same role as the polymer amount concerning the reaching of the saturation of the polymer. The same trend was observed in measuring the surface tension of solutions containing increasing amounts of surfactant and a constant polymer concentration. Referring to Fig. 4, the only case where the polymer saturation seems to be reached is for PEO 1 × 105 Mw ; increasing the chain length above this value, the second transition point in the plot was not observed. Moreover the c.a.c. also seemed decreased when the polymer chain length increased. Also, in this case, polymer molecular weights below 35 × 103 seemed not to induce any surfactant aggregation, even at very high polymer concentration (data not reported in the figure); the only effect observed was a decrease in the surface tension in the very dilute range, due to an additive effect. Considering that the amount of the polymer and its chain length had the same effect (of enhancing the complexation and increasing the surfactant concentration required to form ordinary micelles) and that the phenomenon was not observed for lower chain lengths, our results suggest the existence of a characteristic number of EO units to form the complex. Preliminary experiments on the efficiency of PEO-based retention systems in the presence of fatty acids showed the impairing effect of the latter on the total and fines retention. Moreover, water drainage through the fiber web was affected, too.
4. Conclusions Our experiments showed an interaction between the octadecylcarboxylates with different degrees of unsaturation and the PEO. Even if additional work would be required, the surface tension method was useful enough as an indicator of the polymer–surfactants interactions. The increase in the solubility of the fatty acid seems to lead to a less pronounced interaction, while, in contrast, the increase in the PEO chain length seems to emphasize the phenomenon. The
The authors thank CTP and CTPi members for their financial support.
References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37]
L. Rahman, C.H. Tay, Tappi J. (1986) 100. D.B. Braun, D. Ehms, Tappi J. (1984) 110. K. Stack, L. Dunn, N. Roberts, Appita 43 (1990) 125. K. Stack, L. Dunn, Appita 48 (1995) 83. K. Stack, L. Dunn, S. Maughan, Appita 48 (1995) 275. B. Alince, T. Van Deven, Tappi J. 80 (1997) 181. S. Tay, Tappi J. 80 (1997) 149. R.H. Pelton, C.H. Tay, L.H. Allen, J. Pulp Paper Sci. 10 (1984) J5. M. Polverari, L. Allen, B. Sitholé, P. Gagnon, J.F. Samuel, Tappi J. 84 (2001) 1. M. Polverari, L. Allen, B. Sithole, Tappi J. Peer Reviewed 84 (2001) 1. G. Laivins, M. Polverari, H. Allen, J. Pulp Paper Sci. 27 (2001) 190. R. Haynes, H. Marcoux, in: Proceedings of the CPPA 4th Research Forum on Recycling, Quebec, Canada, 1997, p. 163. D. MacNell, A. Roring, H. Hoel, in: Proceedings of PIRA 6th International Recycling Technology Conference, Hungary, 2000. S.R. Tremont, Appita 1 (1996) 33. S.R. Tremont, in: 81st Annual Meeting Technical Section, Montreal, Canada, 1995, p. 25. R. Ekman, B. Holmbom, Nord. Pulp Paper Res. J. 1 (1989) 16. M.N. Jones, J. Colloid Interface Sci. 23 (1967) 36. M.N. Jones, J. Colloid Interface Sci. 26 (1968) 532. K. Shirahama, Colloid Polym. Sci. 252 (1974) 978. K. Shirahama, N. Ide, J. Colloid Interface Sci. 54 (1976) 450. B. Cabane, R. Duplessix, J. Phys. 43 (1982) 1529. M.J. Schwuger, J. Colloid Interface Sci. 43 (1973) 491. W. Brown, J. Fundin, M. Miguel, Macromolecules 25 (1992) 7192. J.M. Meglio, P. Baglioni, J. Phys. Chem. 98 (1994) 5478. Y. Touhami, D. Rana, G.H. Neale, V. Hornof, Colloid Polym. Sci. 279 (2001) 297. C. Lima, F. Nome, D. Zanette, J. Colloid Interface Sci. 189 (1997) 174. M.I. Gjerde, W. Nerdal, H. Hoiland, J. Colloid Interface Sci. 197 (1998) 191. D. Dhara, D.O. Shah, Langmuir 17 (2001) 7233. L.M. Smitter, J.F. Guedez, A.J. Müller, A.E. Sáez, J. Colloid Interface Sci. 236 (2001) 343. N. Kamenka, R. Zana, J. Colloid Interface Sci. 188 (1997) 130. M.I. Gjerde, W. Nerdal, H. Høiland, J. Colloid Interface Sci. 183 (1996) 285. D.J. Cooke, C.C. Dong, J. Lu, R.K. Thomas, J. Phys. Chem. 102 (1998) 4912. D.J. Cooke, J.A.K. Blondel, J. Lu, R.K. Thomas, Langmuir 14 (1998) 1990. Y. Wang, B. Han, H. Yan, D.J. Cooke, J. Lu, R.K. Thomas, Langmuir 14 (1998) 6054. J.T.C. Kwak, Polymer–Surfactant Systems, Dekker, New York, 1998. E.D. Goddard, J. Colloid Interface Sci. 256 (2002) 228. R. Sartori, L. Sepulveda, F. Quina, E. Lissi, E. Abuin, Macromolecules 23 (1990) 3878.
220
E. Zeno et al. / Journal of Colloid and Interface Science 277 (2004) 215–220
[38] A. Blokhus, K. Klokk, J. Colloid Interface Sci. 230 (2000) 448. [39] D. Zanette, V. Soldi, A. Romani, M.H. Gehlen, J. Colloid Interface Sci. 246 (2002) 387. [40] D. Beneventi, B. Carre, A. Gandini, J. Colloid Interface Sci. 237 (2001) 142.
[41] [42] [43] [44]
J.E. Glass, J. Phys. Chem. 72 (1968) 4459. B.H. Cao, M.W. Kim, Faraday Discuss. 98 (1994) 245. I.P. Purcell, J.R. Lu, R.K. Thomas, Langmuir 14 (1998) 1637. H. Arai, M. Murata, K. Shinoda, J. Colloid Interface Sci. 37 (1971) 223–227.
Interactions between poly(ethylene oxide) and fatty acids sodium salts studied by surface tension measurements Elisa Zeno a,b,∗ , Davide Beneventi b , Bruno Carré b a Ecole Française de Papeterie et des Industries Graphiques (INPG), B.P. 65, 38402 St.-Martin d’Hères, France b Centre Technique du Papier, Ressources Fibreuses, Domaine Universitaire, B.P. 251, 38044 Grenoble Cedex 9, France
Received 20 January 2004; accepted 16 April 2004 Available online 19 May 2004
Abstract This work focuses onto the interactions between poly(ethylene oxide) (PEO) and fatty acids, in order to set their potential role of contaminants for PEO-based retention systems. Surface tension measurements were used to investigate PEO–fatty acid systems and they made it possible to clearly point out the interactions between the polymer and the sodium octadecylcarboxylates with different degrees of unsaturation. The observed interaction seems to be dependent on the fatty acids’ solubility, the increase of which leads to less pronounced phenomena, which are, in contrast, emphasized by the increase in PEO chain length. 2004 Elsevier Inc. All rights reserved. Keywords: Polyethylene oxide; Fatty acids; Surface tension; Degree of unsaturation
1. Introduction The papermaking process involves the use of different polymeric materials aimed to improve the fines and fillers retention in the fiber network. Retention aids can be either polyelectrolytes or nonionic polymers, such as the polyethylene oxide (PEO), which has been found to be an efficient retention aid if used in combination with an enhancer (usually a phenolic resin) [1–7]. During the past years, this system has been successfully applied with mechanical and recycled pulps, for which the conventional cationic retention systems failed because of the high levels of dissolved and colloidal substances [8–10]. Actually, cationic polymers interact with anionic substances originating from wood or carried over from the de-inking process, so that they are unavailable to develop their action onto the fibers. A nonionic retention aid such as PEO, being less sensitive to negative charges, should be able to maintain good retention performance even in the presence of high levels of anionic interfering substances. Nevertheless, several chemicals have been found to interfere with this system, in particular some of the nonionic * Corresponding author. Fax: +33-0-4-76-15-40-16.
E-mail address: [email protected] (E. Zeno). 0021-9797/$ – see front matter 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.jcis.2004.04.028
surfactants used in the de-inking lines and carried over to the paper machine [11]. Provided that the main residual de-inking chemicals carried over to the paper machine are the fatty acids [12–15] and that considerable amounts of these wood extractives can be dispersed or dissolved in the mechanical pulp after the mechanical fibrillation and bleaching [16], it seems important to investigate their interaction with the PEO, in order to set their potential role of contaminants for this retention system. Polymer–surfactant aqueous systems have been widely investigated over a long period with varied experimental techniques; most of these studies dealt with a neutral polymer and an anionic surfactant. In particular, the system of poly(ethylene oxide) (PEO) and sodium dodecyl sulfate (SDS) has been extensively investigated [17–34] and reviewed [35,36]. By means of classical physical and spectroscopic methods, the spontaneous association of the SDS and the PEO has been explained in terms of cooperative binding of the surfactant to the polymer in the form of aggregates, driven by hydrophobic and electrostatic interactions. According to the models proposed, the wrapping of the polymer around micelles would favor micellization, which starts at a concentration (critical aggregation concentration, c.a.c. or c1 ) lower than the critical micellar concentration (c.m.c.) and goes on up to the point of saturation of the polymer
216
E. Zeno et al. / Journal of Colloid and Interface Science 277 (2004) 215–220
(p.p.s. or c2 ). Moreover, the PEO capacity of binding the alkalimetal cations [37] would result in interaction with the counterion, leading to a positive charge distribution which interacts with the anionic surfactant. Referring to these models, some studies deal with the counterion effect on PEO–anionic surfactant complexation [32,33], whereas some others concern the effect of the alkyl chain length [19,20,38] and consequently of the surfactant hydrophobicity. More recently, the effect of the polar head group has been taken into account and the role of the carboxylate head group has been investigated [39], pointing out the stronger binding of the PEO with sodium dodecanoate. Furthermore, a recent study [38] reports the decrease in the PEO–sodium alkyl carboxylate interactions when the hydrophobicity of the surfactants is varied by decreasing the chain length from C12 to C8, where no interaction was observed. Actual results and the previous considerations of PEO use as a retention aid promoted this work, aimed at investigating the interaction between PEO and sodium octadecyl carboxylates with different degrees of unsaturation, and then focusing on the oleic acid, which is reported to be one of the dominant fatty acids in a mechanical pulp sample [16].
respectively the surfactant concentration, at a value above the c.m.c., and the polymer concentration, at 1 × 10−6 M. For the sodium oleate, some measurements were taken, keeping constant the molar ratio between PEO and NaOl. An UV/visible spectrophotometer was used to measure the light transmission of PEO–fatty acid systems. Measurements were performed at room temperature. The transmittance values at 260 nm were used to calculate a lightscattering coefficient according to Lambert’s law, I /I0 = e−γ x ,
(1)
where I is the intensity of the incident light, I0 the intensity of the incident light after passing through a thickness x of the liquid sample, and γ the extinction coefficient, which is composed of two terms, the scattering coefficient (τ ) and the absorption coefficient (κ), γ = τ + κ.
(2)
When light scattering is large compared to absorption, as in the case of PEO and fatty acid systems, the absorption coefficient can be neglected and Eq. (1) becomes I /I0 = e−τ x . The transmittance (T = I /I0 ) values are related to the scattering coefficient by the equation T = I /I0 = e−τ x ,
(3)
which was used to calculate τ .
2. Experimental 2.1. Materials and methods
3. Results and discussion High-purity commercial samples of poly(ethylene oxide) and poly(ethylene glycol) with average molecular weights of 194, 35 × 103, 1 × 105 , 6 × 105 , 9 × 105 , and 5 × 106 were used in this study. Sodium stearate (NaSt), sodium oleate (NaOl), and linolenic acid (NaLin) were all high purity grades (minimum 99%) and used as received. All solutions were prepared at pH 10 with concentrated commercial buffer solution (boric acid/potassium chloride/sodium hydroxide) and water purified with a Milli-Q unit. Linolenic acid dissolved at pH 10 in the presence of NaOH forms the corresponding sodium salt, so it is labeled (NaLin). Equilibrium surface tension measurements were performed with a De Noüy ring apparatus at 35 or 70 ◦ C. The adsorption isotherms of surfactants and polymers of different molecular weight were determined by measuring the surface tension value 3 min after the addition of increasing volumes of a concentrated mother solution to a fixed volume of deionized water. The ternary systems were investigated, keeping constant
3.1. Binary systems 3.1.1. Fatty acid sodium salts Adsorption isotherms showed that the increase in the fatty acid chain unsaturation led to an increase in the minimum surface tension and in the c.m.c. value (see Table 1). This behavior, in agreement with the findings of a previous work [40], can be interpreted as a result of the increase in hydrophilicity of the fatty acids, due to the chain unsaturation increase. 3.1.2. Poly(oxyethylene) Fig. 1 shows the results obtained for aqueous solutions of polyoxyethylene. For lower molecular-weights, the surface tension exhibits a strong molecular-weight dependence, while, for higher molecular weights, where the hydroxyl end group contributions become insignificant, all the samples
Table 1 Main properties of the fatty acids used in this study Fatty acid Stearic Oleic Linolenic
Formula CH3 –(CH2 )16 –CO2 H CH3 (CH2 )7 CH=CH(CH2 )7 CO2 H CH3 (CH2 CH=CH)3 (CH2 )7 CO2 H
Mp (◦ C) 69.6 13.4 −5
a As obtained from adsorption isotherms at pH 10 and 35 ◦ C (70 ◦ C for the stearic acid).
c.m.c.a (M) 7 × 10−5 7 × 10−5 7 × 10−4
γmin a (mN/m) 25.5 28 32.5
E. Zeno et al. / Journal of Colloid and Interface Science 277 (2004) 215–220
217
Table 2 Surface tension rise of the fatty acids micellar solution, subsequent to the addition of PEO Fatty acid
No. double bonds
γ − γmin (mN/m) 70 ◦ C 35 ◦ C
Stearic Oleic Linolenic
C (18:0) C (18:1) C (18:3)
26.9 11.1 –b
–a 20.5 4.2
a T ◦ Krafft = 66 C. b Difference within experimental error.
Fig. 1. Surface activity–molecular weight dependence of polyoxyethylene aqueous solutions (35 ◦ C).
Fig. 3. Schematic surface tension/log concentration plot of a surfactant (lower dotted line) and of a surfactant in the presence of a complexing polymer (full line: idealized behavior and dotted lines: experimental behavior) according to [36].
Fig. 2. Effect of the PEO (Mw = 9 × 105 ) addition on a micellar solution of NaSt at 70 ◦ C (F), NaOl at 70 ◦ C (P) and at 35 ◦ C (Q), and NaLin at 35 ◦ C (!).
have the same behavior, i.e., a region where the surface tension (60–61 mN/m) no longer decreases increasing the polymer concentration. In accord with several findings reported in the literature [41,42], the stabilization of the surface tension of high-Mw PEO solutions was associated with the saturation of the air/water interface by the polymer chains. 3.2. Ternary systems When PEO (9 ×105 Mw ) was added to a micellar solution of each fatty acid, the surface tension increased when the polymer concentration increased. The shape of the surface tension versus PEO concentration curve shown in Fig. 2 led to the assumption that on adding low amounts of polymer to the soap solution, a soluble complex was formed and, when a certain polymer/surfactant ratio was reached, phase separation occurred. The increase in surface tension was therefore correlated with the decrease in the surfactant concentration in the aqueous phase. It should be noted that measurements were performed at a surfactant concentration slightly above the c.m.c., in order to avoid dilution effects due to the addition of the polymer. Moreover, in the range of concentrations tested, the polymer surface tension was constant at the plateau value (60–61 mN/m). The PEO–surfactant binding
seems to be weaker when the number of double bonds increases, as shown in Table 2. This result is in line with the finding of a recent work [38], where it was shown that decreasing the chain length of the sodium alkylcarboxylates, the polymer–surfactant interactions becomes weaker. Actually, both the decrease in the number of carbon atoms and the increase in the number of double bonds result in a decrease in the surfactant hydrophobicity, which is an important parameter governing polymer–surfactant complexation. It should be noted that measurements were carried out at pH 10, but, as preliminary experiments seem to indicate, results might be applicable to lower alkaline pH, given that fatty acids are under the dissociated form for pH > 7; as an example, pK = 4.95 for the oleic acid. Provided that (i) PEO forms complexes with fatty acid sodium soaps, (ii) the strength of the interaction is weaker when the degree of unsaturation increases, i.e., with the linolenic acid, and (iii) in paper mill process water the stearic acid is totally precipitated because of temperature (below the Krafft point) and because of the presence of calcium ions, a more detailed investigation of the polymer–surfactant interaction was carried out only for the sodium oleate. A common method for studying the polymer–surfactant interaction is surface tension measurement as a function of surfactant concentration at a constant polymer concentration. The classical result [36], if complexation occurs, is schematically represented in Fig. 3. The first breaking point is the T1 or c.a.c., where the interaction starts and the surfactant, which was present only as monomers up to the c.a.c., begins to micellize onto the polymer in the bulk. In the dilute range, up to the c.a.c., the
218
E. Zeno et al. / Journal of Colloid and Interface Science 277 (2004) 215–220
Fig. 4. Surface tension of solutions containing a constant PEO concentration (1 × 10−6 M) and increasing NaOl concentrations.
surface tension is always below that of the surfactant solution itself, because of the polymer surface activity or of its enhancing effect on the surfactant adsorption in the case of large attractive interactions between the two components (see the poly(vynilpyrrolidone) (PVP)-SDS system [43]). At the second transition, T2 , the polymer is saturated and there is an accumulation of surfactant monomers, resulting in a surface tension decrease up to T2 , where ordinary aqueous surfactant micelles are formed. In such a model the surface tension behavior could be explained by changes in the bulk phase and the polymer/surfactant complex has no surface activity. In our conditions, PEO concentration fixed at 1 × 10−6 M, only the first break point was observed (Fig. 4) with a progressive shift of T1 toward low sodium oleate concentrations with increasing PEO Mw , i.e., T1 = 1.8 × 10−5 M for PEO 35,000 and T1 = 2 × 10−6 for PEO 900,000, which was associated with an increase in the surfactant–polymer interactions with the PEO Mw . Considering that the surfactant concentration at the T2 point is influenced by the polymer concentration [22], the reason could be a too high polymer amount to reach saturation in the concentration range tested. This hypothesis is in agreement with the dependence of the concentration interval between T1 and T2 on several parameters such as pH, T ◦ , counterions and polymer amount, reported in the literature [22,44]. In order to investigate the effect of the PEO amount, further measurements were carried out, keeping constant the PEO/NaOl molar ratio and varying the ratio between 0.0001 and 0.1. The results obtained are shown in Fig. 5 and it could be seen that for a surfactant concentration above the c.m.c. the saturation point would be reached only for the lowest ratio, which corresponds to a polymer concentration of 2 × 10−8 M. Moreover, by increasing the ratio PEO/NaOl the complexation phenomenon occurred for lower surfactant concentration, indicating that the polymer amount plays also a role on the beginning of the complexation. The complex formation was confirmed also by the light transmission measurements, according to Eq. (3). The scattering coefficient of NaOl–PEO mixtures was calculated and plotted versus surfactant concentration (Fig. 6). The NaOl concentration at the breaking point is the same as the concentration at the break-
Fig. 5. Effect of the PEO/NaOl molar ratio on the surface tension of NaOl (Q) in the absence of PEO and for PEO/NaOl mixtures (E, 1, !).
Fig. 6. Scattering coefficient as calculated from Eq. (3) for PEO/NaOl mixtures.
Fig. 7. Effect of PEO addition on a micellar solution of NaOl, varying the polymer chain length.
point in the surface tension versus concentration plot shown in Fig. 5. The interaction between NaOl and PEO having been verified, the effect of the PEO amount, we investigated the influence of the PEO chain length on the complexation, from the tetrameric species to a PEO of average molecular weight 5 × 106. Their addition to a NaOl micellar solution gave the results illustrated in Fig. 7. The addition of the oligomers did not induce any change in the surface tension, even when
E. Zeno et al. / Journal of Colloid and Interface Science 277 (2004) 215–220
219
binding of the surfactants to the polymer could be at the base of the impairing effect of the fatty acids on the retention system constituted by PEO and phenol formaldehyde resin, observed in preliminary work.
Acknowledgment Fig. 8. Molar ratio PEO/NaOl at which is observed the rise in the surface tension of a micellar solution of NaOl when increasing amounts of PEO are added.
the oligomer–surfactant ratio was hugely increased (data not reported in the figure). At higher molecular weights, the surface tension rose and the polymer concentration at which the increase started was dependent on the PEO chain length. In Fig. 8 are reported the molar ratios between PEO and NaOl at which the surface tension increase began, as a function of the PEO chain length. It could be assumed that the chain length plays the same role as the polymer amount concerning the reaching of the saturation of the polymer. The same trend was observed in measuring the surface tension of solutions containing increasing amounts of surfactant and a constant polymer concentration. Referring to Fig. 4, the only case where the polymer saturation seems to be reached is for PEO 1 × 105 Mw ; increasing the chain length above this value, the second transition point in the plot was not observed. Moreover the c.a.c. also seemed decreased when the polymer chain length increased. Also, in this case, polymer molecular weights below 35 × 103 seemed not to induce any surfactant aggregation, even at very high polymer concentration (data not reported in the figure); the only effect observed was a decrease in the surface tension in the very dilute range, due to an additive effect. Considering that the amount of the polymer and its chain length had the same effect (of enhancing the complexation and increasing the surfactant concentration required to form ordinary micelles) and that the phenomenon was not observed for lower chain lengths, our results suggest the existence of a characteristic number of EO units to form the complex. Preliminary experiments on the efficiency of PEO-based retention systems in the presence of fatty acids showed the impairing effect of the latter on the total and fines retention. Moreover, water drainage through the fiber web was affected, too.
4. Conclusions Our experiments showed an interaction between the octadecylcarboxylates with different degrees of unsaturation and the PEO. Even if additional work would be required, the surface tension method was useful enough as an indicator of the polymer–surfactants interactions. The increase in the solubility of the fatty acid seems to lead to a less pronounced interaction, while, in contrast, the increase in the PEO chain length seems to emphasize the phenomenon. The
The authors thank CTP and CTPi members for their financial support.
References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37]
L. Rahman, C.H. Tay, Tappi J. (1986) 100. D.B. Braun, D. Ehms, Tappi J. (1984) 110. K. Stack, L. Dunn, N. Roberts, Appita 43 (1990) 125. K. Stack, L. Dunn, Appita 48 (1995) 83. K. Stack, L. Dunn, S. Maughan, Appita 48 (1995) 275. B. Alince, T. Van Deven, Tappi J. 80 (1997) 181. S. Tay, Tappi J. 80 (1997) 149. R.H. Pelton, C.H. Tay, L.H. Allen, J. Pulp Paper Sci. 10 (1984) J5. M. Polverari, L. Allen, B. Sitholé, P. Gagnon, J.F. Samuel, Tappi J. 84 (2001) 1. M. Polverari, L. Allen, B. Sithole, Tappi J. Peer Reviewed 84 (2001) 1. G. Laivins, M. Polverari, H. Allen, J. Pulp Paper Sci. 27 (2001) 190. R. Haynes, H. Marcoux, in: Proceedings of the CPPA 4th Research Forum on Recycling, Quebec, Canada, 1997, p. 163. D. MacNell, A. Roring, H. Hoel, in: Proceedings of PIRA 6th International Recycling Technology Conference, Hungary, 2000. S.R. Tremont, Appita 1 (1996) 33. S.R. Tremont, in: 81st Annual Meeting Technical Section, Montreal, Canada, 1995, p. 25. R. Ekman, B. Holmbom, Nord. Pulp Paper Res. J. 1 (1989) 16. M.N. Jones, J. Colloid Interface Sci. 23 (1967) 36. M.N. Jones, J. Colloid Interface Sci. 26 (1968) 532. K. Shirahama, Colloid Polym. Sci. 252 (1974) 978. K. Shirahama, N. Ide, J. Colloid Interface Sci. 54 (1976) 450. B. Cabane, R. Duplessix, J. Phys. 43 (1982) 1529. M.J. Schwuger, J. Colloid Interface Sci. 43 (1973) 491. W. Brown, J. Fundin, M. Miguel, Macromolecules 25 (1992) 7192. J.M. Meglio, P. Baglioni, J. Phys. Chem. 98 (1994) 5478. Y. Touhami, D. Rana, G.H. Neale, V. Hornof, Colloid Polym. Sci. 279 (2001) 297. C. Lima, F. Nome, D. Zanette, J. Colloid Interface Sci. 189 (1997) 174. M.I. Gjerde, W. Nerdal, H. Hoiland, J. Colloid Interface Sci. 197 (1998) 191. D. Dhara, D.O. Shah, Langmuir 17 (2001) 7233. L.M. Smitter, J.F. Guedez, A.J. Müller, A.E. Sáez, J. Colloid Interface Sci. 236 (2001) 343. N. Kamenka, R. Zana, J. Colloid Interface Sci. 188 (1997) 130. M.I. Gjerde, W. Nerdal, H. Høiland, J. Colloid Interface Sci. 183 (1996) 285. D.J. Cooke, C.C. Dong, J. Lu, R.K. Thomas, J. Phys. Chem. 102 (1998) 4912. D.J. Cooke, J.A.K. Blondel, J. Lu, R.K. Thomas, Langmuir 14 (1998) 1990. Y. Wang, B. Han, H. Yan, D.J. Cooke, J. Lu, R.K. Thomas, Langmuir 14 (1998) 6054. J.T.C. Kwak, Polymer–Surfactant Systems, Dekker, New York, 1998. E.D. Goddard, J. Colloid Interface Sci. 256 (2002) 228. R. Sartori, L. Sepulveda, F. Quina, E. Lissi, E. Abuin, Macromolecules 23 (1990) 3878.
220
E. Zeno et al. / Journal of Colloid and Interface Science 277 (2004) 215–220
[38] A. Blokhus, K. Klokk, J. Colloid Interface Sci. 230 (2000) 448. [39] D. Zanette, V. Soldi, A. Romani, M.H. Gehlen, J. Colloid Interface Sci. 246 (2002) 387. [40] D. Beneventi, B. Carre, A. Gandini, J. Colloid Interface Sci. 237 (2001) 142.
[41] [42] [43] [44]
J.E. Glass, J. Phys. Chem. 72 (1968) 4459. B.H. Cao, M.W. Kim, Faraday Discuss. 98 (1994) 245. I.P. Purcell, J.R. Lu, R.K. Thomas, Langmuir 14 (1998) 1637. H. Arai, M. Murata, K. Shinoda, J. Colloid Interface Sci. 37 (1971) 223–227.