Interactions of Poly(amidoamine) Dendrimers with the Surfactants

Christopher V. Kelly , Meghan G. Liroff , L. Devon Triplett , Pascale R. Leroueil , Douglas G. Mullen , Joseph M. Wallace , Sasha Meshinchi , James R...
0 downloads 0 Views 205KB Size
Download PDF
9320

Langmuir 2004, 20, 9320-9328

Interactions of Poly(amidoamine) Dendrimers with the Surfactants SDS, DTAB, and C12EO6: An Equilibrium and Structural Study Using a SDS Selective Electrode, Isothermal Titration Calorimetry, and Small Angle Neutron Scattering J. Sidhu,† (the late) D. M. Bloor,† S. Couderc-Azouani,‡ J. Penfold,§ J. F. Holzwarth,*,‡,| and E. Wyn-Jones*,† School of Chemical Sciences, Science Research Institute, University of Salford, Salford M5 4WT, U.K., Fritz-Haber-Institut der Max-Planck-Gesellschaft, Faradayweg 4-6, D-14195 Berlin (Dahlem), Germany, and ISIS Facility, Rutherford Appleton Laboratory, Chilton Didcot, Oxfordshire OX11 O QX, U.K. Received February 26, 2004. In Final Form: July 30, 2004 Interactions in aqueous solutions of different generations of poly(amidoamine) (PAMAM) dendrimers containing amine, hydroxyl, or δ-glucolactone functional groups at the periphery with the anionic surfactant sodium dodecyl sulfate (SDS) were investigated. We used a SDS-specific electrode (EMF) for SDS monomer concentration monitoring, isothermal titration calorimetry (ITC) for binding information, and small angle neutron scattering (SANS) for structural studies. ITC experiments monitoring the interaction of the dendrimers with cationic dodecyltrimethylammonium bromide (DTAB) and nonionic hexaethylene glycol mono-n-dodecyl ether (C12EO6) showed no significant binding effects. In contrast, SDS binds to all of the above dendrimers. EMF and ITC data demonstrated a regular trend for both the onset of binding and binding saturation as the generation in each family of dendrimers increased. In addition, generation G6 exhibited a noncooperative binding process at very low SDS concentrations. Furthermore, the onset of cooperative binding in the EMF experiments started at lower concentrations as the weight % (w/v), the size, and the numbers of the internal or surface groups increased. On the other hand, the binding capacity of the dendrimers showed only a small dependence on the above parameters. At SDS concentrations approaching the binding limit and also at selective concentrations within the binding range, SANS measurements indicated that in all cases the bound surfactant is in the micellar form. From the electromotive force (EMF) measurements, ITC data, and SANS data, the stoichiometry of the supramolecular complexes was determined.

Introduction 1

Starburst dendrimers are a class of highly branched polymers synthesized from various initiator cores via covalently bonded layers generation by generation resulting in macromolecules with well-defined radial branches, very specific molecular masses, and uniform sizes.2,3 Over the past 20 years, dendrimers have found wide application in many commercial areas, for example, in ink formulation, personal care products, chemical sensors, controlled agent release, and the reclamation of catalysts.4 The polymer terminates in a radially templated surface with a high number of accessible reactive groups, which may be given defined functionality to moderate interaction and provide a degree of specificity to the interaction of the dendrimer with small molecules. Unlike classical polymers, dendrimers have a high degree of molecular uniformity, with a moderately hydrophobic interior and a highly functional hydrophilic * Corresponding authors. † University of Salford. ‡ Fritz-Haber-Institut der Max-Planck-Gesellschaft. § Rutherford Appleton Laboratory. | Phone: +49 30 8413 55 16. Fax: +49 30 84 13 53 85. E-mail: [email protected]. (1) Flory, P. J. J. Am. Chem. Soc. 1941, 63, 3083, 3091, and 3096. (2) Tomalia, D. A.; Baker, H.; Dewald, J.; Hall, M.; Kallos, G.; Martin, S.; Roeck, J.; Ryder, J.; Smith, P. Polymer J. 1985, 17, 117. (3) Tomalia, D. A.; Baker, H.; Dewald, J.; Hall, M.; Kallos, G.; Martin, S.; Roeck, J.; Ryder, J.; Smith, P. Macromolecules 1986, 19, 2466. (4) Advances in Dendritic Macromolecules; Newkome, G. R., Ed.; JAI Press: Greenwich, CT, 1993.

terminal surface. Host-guest chemistry5 can take place either at the interior or at the periphery of the dendrimer.5,6 Surfactants make ideal compounds for such studies because of the variety available in their chemical structure allowing for systematic investigations. The binding of surfactants to dendrimers to form supramolecular assemblies was initially investigated by Turro and Tomalia using probe molecules whose electron spin resonance (ESR)7,9 or fluorescence spectra6,7 are sensitive to the microenvironment in which they reside. The spectral characteristics of these probe molecules have been used to demonstrate that dendrimers (with carboxylic groups at the surface) promote the binding of surfactants and to detect the aggregates of bound surfactants (cationic surfactants). A tentative structural model was proposed in 1997 by Turro et al.9,10 Esumi et al.11-13 also studied the interaction of sodium dodecyl sulfate (SDS) with poly(amidoamine) dendrimers, their sugar-persubstituted (5) (a) Zimmerman, S. C.; Lawless, L. J. Top. Curr. Chem. 2001, 217, 95. (b) Esfand, R.; Tomalia, D. A. Drug Discovery Today 2001, 6, 427. (c) Aulenta, F.; Hayes, W.; Rannard, S. Eur. Polym. J. 2003, 39, 1741. (6) Watkins, D. M.; Sayed-Sweet, Y.; Klimash, J. W.; Turro, N. J.; Tomalia, D. A. Langmuir 1997, 13, 3136. (7) Ottaviani, M. F.; Turro, N. J.; Jockusch, S.; Tomalia, D. A. Colloids Surf., A 1996, 115, 9. (8) Caminati, G.; Turro, N.; Jockusch, S.; Tomalia, D. A. J. Am. Chem. Soc. 1990, 112, 8515. (9) Ottaviani, M. F.; Daddi, R.; Brustolon, M.; Turro, N. J.; Tomalia, D. A. Appl. Magn. Reson. 1997, 13, 347. (10) Ottaviani, M. F.; Andechaga, P.; Turro, N. J.; Tomalia, D. A. J. Phys. Chem. B 1997, 101, 6057. (11) Miyazaki, M.; Torigoe, K.; Esumi, K. Langmuir 2000, 16, 1522.

10.1021/la0494932 CCC: $27.50 © 2004 American Chemical Society Published on Web 09/04/2004

Interactions of PAMAM Dendrimers with Surfactants

derivatives, and a hydrophobically modified dendrimer using surface tension and pyrene (probe) fluorescence spectroscopy. The existence of dendrimer/micellar SDS complexes was inferred from the experimental data. It could be argued however that spectroscopic measurements using probes, as done by these authors, as well as surface tension measurements only provide indirect evidence on both structure and equilibrium data. We chose an alternative approach to study the interaction of poly(pyrrolidone-1,4-diaminobutane)14 and poly(1,4-diaminobutane)14,15a dendrimers with sodium dodecyl sulfate (SDS). We used a dodecyl sulfate surfactant selective electrode to provide direct analytical information on the binding of SDS to the dendrimers via free monomer concentration evaluation and isothermal titration calorimetry to monitor detailed changes of interaction enthalpy during the binding process. Finally, small angle neutron scattering was used to provide structural information.14,15a This article is a continuation of our earlier studies on surfactant/polymer interactions14,15 and describes the use of a surfactant ion-specific electrode to determine the binding isotherm and isothermal titration calorimetry to monitor the interaction enthalpy of poly(amidoamine) (PAMAM) dendrimers having different surface groups with sodium dodecyl sulfate (SDS). To get a handle on the influence of different surface groups in affecting the binding process, we used amine, hydroxyl, and δ-glucolactone sugar groups. In addition, small angle neutron scattering was applied to investigate the structures of the surfactant/dendrimer complexes. Experimental Section All the dendrimers had an ethylene 1,2-diamine core and various generations of commercially available poly(amidoamine) in the following layers with surface amino groups code-named Gx-PAMAM-NH2 (see Chart 1 for G4) for generations 3-6 (G3G6) or with surface OH functional groups code-named GxPAMAM-OH (G3-G6). The commercial samples (10% in methanol) were purchased from Aldrich; the methanol was evaporated in a vacuum oven at 333 K for several hours, and the resulting solid was used without further purification because NMR and mass spectrometry showed no impurities. We synthesized some compounds ((G3-G6)-PAMAM-NH2 and G3PAMAM-OH) and checked their NMR spectra, which agreed very well with the commercial samples. Furthermore, we obtained the same isothermal titration calorimetry (ITC) results for both the commercial and synthesized samples. We also synthesized and purified sugar-ball dendrimers with δ-glucolactone functional groups at the periphery of the dendrimers (code-named GxPAMAM-sugar ball (G3-G5)) according to procedures described in the literature.13,16 Sodium dodecyl sulfate (SDS) was purchased from Sigma-Aldrich and purified by recrystallization with analar ethanol (three times). Dodecyltrimethylammonium bromide (DTAB) was obtained from Sigma-Aldrich and purified by repeated recrystallization from ethanol, and hexaethylene glycoln-dodecyl ether (C12EO6) was purchased from Fluka and used as received. The experimental methods used in this study were as follows. (12) (a) Esumi, K.; Saika, R.; Miyazaki, M.; Torigoe, K.; Koide, Y. Colloids Surf., A 2000, 166, 115. (b) Esumi, K.; Kuwabara, K.; Chiba, T.; Kobayashi, F.; Mizutani, H.; Torigoe, K. Colloids Surf., A 2002, 197, 141. (13) Esumi, K.; Chiba, T; Mizutani, H; Shoji, K.; Torigoe, K. Colloids Surf., A 2001, 179, 103. (14) Ghoreishi, S. M.; Li, Y.; Holzwarth, J. F.; Khoshdel, E.; Warr, J.; Bloor, D. M.; Wyn-Jones, E. Langmuir 1999, 15, 1938. (15) (a) Li, Y.; McMillan, C. A.; Bloor, D. M.; Penfold, J.; Warr, J.; Holzwarth, J. F.; Wyn-Jones, E. Langmuir 2000, 16, 7999. (b) Li, Y.; Xu, R.; Couderc, S.; Ghoreishi, S. M.; Warr, J.; Bloor, D. M.; Holzwarth, J. F.; Wyn-Jones, E. Langmuir 2003, 19, 2026. (c) Ghoreishi, S. M.; Li. Y.; Bloor, D. M.; Warr, J.; Wyn-Jones, E. Langmuir 1999, 15, 4380. (16) (a) Schmitzer, A.; Perez, E.; Rico-Lattes, I.; Lattes, A.; Rosca, S. Langmuir 1999, 15, 4397. (b) Aoi, K.; Itoh, K.; Okada, M. Macromolecules 1995, 28, 5391.

Langmuir, Vol. 20, No. 21, 2004 9321 Chart 1. Generation 4 Structure of the PAMAM-NH2 Dendrimer (Poly(amidoamine) Dendrimer with Amino Groups on the Surface)

(i) Electromotive Force (EMF) Measurements. The surfactant selective membrane electrodes applied in the present work were fabricated using procedures which were described previously.17-19 The membrane was constructed from specially conditioned poly(vinyl chloride) (PVC) and a commercially available polymeric plasticizer. All solutions were doped with NaBr (10-4 mol dm-3), allowing the EMF of the surfactant electrode to be measured relative to a commercial bromide ion electrode both for pure SDS and also SDS in the presence of PAMAM dendrimers of various generations. The measurements were taken at 298 K. In practice, the EMF data monitor the monomer surfactant concentration (m1) as a function of total added surfactant (C), and when aggregation occurs, the difference (C - m1) represents the amount of bound surfactant. (ii) Isothermal Titration Calorimetry (ITC). The isothermal titration microcalorimeter used was the MicroCal Omega ITC instrument from MicroCal Inc., Northampton, MA In ITC experiments, one measures directly the enthalpy changes associated with processes occurring at constant temperature. Experiments were carried out by titrating micellar SDS into an aqueous solution of the dendrimer. An injection schedule (number of injections, 50; volume of injections, 10 µL; time between injections, 4 min) was set up using interactive software, and this schedule was automatically carried out with all data stored to disk. After each addition, the heat released or absorbed as a result of the various interaction processes occurring in the solution mixture was monitored by the microcalorimeter.20,21 (iii) Small Angle Neutron Scattering. (a) Experimental. The small angle neutron scattering (SANS) measurements were made on the LOQ diffractometer22 of the ISIS pulsed neutron source at the Rutherford Appleton Laboratory. The measurements were made using the white-beam time-of-flight method to give a Q range of 0.02-0.15 Å-1. A beam aperture of 12 mm and a sample path length of 5 mm were used, and the samples were measured at a temperature of 298 K. The instrument configuration and sample geometry were optimized to give maximum sensitivity to low surfactant concentrations, over a limited Q range. The data were corrected for background scattering and (17) Painter, D. M.; Bloor, D. M.; Takisawa, N.; Hall, D. G.; WynJones, E. J. Chem. Soc., Faraday Trans. 1 1988, 84, 2087. (18) Takisawa, N.; Brown, P.; Bloor, D. M.; Hall, D. G.; Wyn-Jones, E. J. Chem. Soc., Faraday Trans. 1 1989, 85, 2099. (19) Wan-Badhi, W. A.; Wan-Yunus, W. M. Z.; Bloor, D. M.; Hall, D. G.; Wyn-Jones, E. J. Chem. Soc., Faraday Trans. 1993, 89, 2737. (20) Li, Y.; Xu, R.; Bloor, D. M.; Holzwarth, J. F.; Wyn-Jones, E. Langmuir 2000, 16, 10515. (21) Li, Y.; Xu, R.; Couderc, S.; Bloor, D. M.; Holzwarth, J. F.; WynJones, E. Langmuir 2001, 17, 5742. (22) Heenan, R. K.; Penfold, J.; King, S. M. J. Appl. Crystallogr. 1997, 30, 1140.

9322

Langmuir, Vol. 20, No. 21, 2004

Sidhu et al.

Table 1. Number of Terminal Groups and of Internal Nitrogen Groups for the Different Generations of Dendrimers with an Ethylene 1,2-Diamine Core dendrimer generation G3-PAMAM-X G4-PAMAM-X G5-PAMAM-X G6-PAMAM-X

no. of N no. of NH2, OH, or molecular mass groups in the sugar-ball groups (MM) for Gxinternal core at the periphery (X) PAMAM-NH2 42 90 186 378

16 32 64 128

3256 6909 14215 28825

detector responses and converted to the scattering cross section (in absolute units of cm-1) using standard procedures.23 (b) Theory. The micelle structure was established by analyzing the scattering data using a standard and now well-established model for micelles.24 For a solution of globular polydisperse interacting particles, the coherent scattering cross section can be written by the so-called “decoupling approximation” (assuming that there are no correlations between position, orientation, and size)24,25

dσ/dΩcoh ) Np[S(Q)〈|F(Q)|2〉 + |〈F(Q)〉|2 - 〈|F(Q)|〉2] (1) where the averages denoted by 〈 〉 are averages over particle size and orientation. Np is the particle number density, S(Q), the structure factor, and F(Q), the particle form factor. The micelles are modeled as “core + shell”,24 and hence, the form factor is

F(Q) ) V1(F1 - F2)F0(QR1) + V2(F2 - Fs)F0(QR2)

(2)

where Vi ) (4/3)πRi3, F0(QR) ) 3j1(QR)/(QR), and F1, F2, and Fs are the scattering length densities of the micelle core, the micelle shell, and the solvent, respectively. The inner core, made up of alkyl chains only, is constrained to space fill a volume limited by a radius (R1) and defined by the fully extended chain length of the surfactant. Any remaining part of the alkyl chains (1 alf), headgroups, and corresponding hydration defines the radius of the outer shell (R2). Polydispersity is included using a Schultz size distribution of micelle sizes,25 which is a convenient distribution whose form is closely associated with the accepted theoretical forms of micelle polydispersity. The interparticle interactions are included using the rescaled mean spherical approximation (RMSA) calculation26,27 for a repulsive (or attractive) Yukawa potential. The adjustable model parameters are then the aggregation number (Nagg), surface charge (Zed) (or potential, in units of kT, UD/kT), and polydispersity (poly). The model was convoluted with the known instrument resolution and compared with the data on an absolute intensity scale on a least-squares basis. Acceptable model fits require not only that the shape of the scattering is reproduced but also that the absolute value of the scattering cross section is in agreement, and this is reflected in the value of the scale factor (sf) (data/theory), where an acceptable variation is ∼ (10%.

Results and Discussion In the present work, we treated the interaction of SDS with the PAMAM dendrimers in the context of classical polymer/surfactant binding studies in the sense that we kept the concentration of dendrimer constant and varied the SDS concentration via the titration procedure. Normally, with linear polymers, the polymer concentration is expressed in % w/v. In the present case, the PAMAM dendrimers have well-defined structures (see Table 1) and molecular masses, and therefore, their molar concentrations were used and in all cases this was kept constant at 10-4 mol dm-3. (23) Heenan, R. K.; King, S. M.; Osborn, R.; Stanley, H. B. RAL Intl. Rep. 1989, RAL-89-128. (24) Hayter, J. B.; Penfold, J. Colloid. Polym. Sci. 1983, 261, 1022. (25) Hayter, J. B. In Physics of Amphiphiles, Micelles, Vesicles and Microemulsions; Degiorgio, V., Corti, M., Eds.; North-Holland Publ.: Amsterdam, The Netherlands, 1992. (26) Hayter, J. B.; Penfold, J. Mol. Phys. 1981, 42, 109. (27) Hansen, J. P.; Hayter, J. B. Mol. Phys. 1982, 46, 651.

The majority of these measurements were carried out at the natural pH (see Table 2) of the system as the SDS titration proceeded. The pKa’s of the PAMAM dendrimers were determined at different ionic strengths using pH titrations by Cakara et al.28 for (G0-G6)-PAMAM-NH2 and Niu et al.29 for G4-PAMAM-OH. A close examination of the reported data shows that the pKa’s are slightly dependent on the ionic strength of the solution and the “Z”-shaped curves in which the fraction of positively charged N groups are plotted against pH shift slightly to lower pHs as the ionic strength decreases. Under the same conditions, the pKa’s of the N groups in the internal core are ∼6 and do not seem to vary for different generations or different peripheral groups. The pKa’s of the peripheral primary nitrogen groups are ∼9 and seem fairly constant in the present binding studies. If we examine the data in refs 28 and 29 carefully and allow for the low ionic strength in our binding experiments, we estimate that for all generations ∼3% (maximum) if any of the peripheral N’s are charged. A calculation based on the structure of the dendrimers shows that the number of terminal groups and the number of internal nitrogen groups are as shown in Table 1. These numbers also tie in with binding studies that we are currently conducting at pH 7 where the dendrimers are essentially high charge density polycations. In these EMF binding studies for the G6-PAMAM-NH2 dendrimer/SDS mixture, the amount of bound SDS at the point where all the -N+ is fully neutralized corresponds to the theoretical number of charged nitrogen groups. (For example, for 10-4 M G6-PAMAM-NH2, precipitation occurs at around 13 mM SDS, corresponding to the protonation of 128 NH2 surface groups (publication in progress).) Since the dendrimers are synthesized generation by generation from G0, we have every reason to believe that all the PAMAM dendrimers contain the theoretical amount of N’s. The systematic progression from low to high generation during the ITC binding studies supports that all the dendrimer samples are pure. In theory, when the binding between SDS and the macromolecule takes place, both the ∆Hi and EMF values of the SDS electrode with and without the polymer are different for each corresponding titration. Previous studies showed that the binding data from many polymer/ surfactant systems involving nonionic polymers and SDS exhibit a clear onset and end of binding. These critical concentrations are denoted Conset (or T1)30 and Csat (or T2). Under these circumstances, the onset of cooperative binding is generally accompanied with a sharp break in the EMF and other physical parameters and is associated with the formation of micelles on the polymer. Hence, it is often referred to as the critical aggregation concentration (cac). In reality, Conset (or T1) is equivalent to the critical micelle concentration (cmc) for the surfactant in the presence of polymer. The end of binding corresponds to the merge of the curves with or without polymer at Csat (or T2), where the polymer is fully saturated with bound SDS. After Csat (or T2), free SDS micelles appear in solution in addition to the bound ones. In practice, the occurrence of both Conset (or T1) and Csat (or T2) is fairly clear in the EMF experiments and Csat (or T2) can also be clearly identified in the ITC experiments. On the other hand, the occurrence of a Conset (or T1) value in the ITC experiments is not always obvious, and in some cases, we were unable (28) (a) Cakara, D.; Kleimann, J.; Borkovec, M. Macromolecules 2003, 36, 4201. (b) Kleinman, M. H.; Flory, J. H.; Tomalia, D. A.; Turro, N. J. J. Phys. Chem. B 2000, 104, 11472. (29) Niu, Y.; Sun, L.; Crooks, R. M. Macromolecules 2003, 36, 5725. (30) Jones, M. N. J. Colloid Interface Sci. 1967, 23, 36.

Interactions of PAMAM Dendrimers with Surfactants

Langmuir, Vol. 20, No. 21, 2004 9323

Table 2. Conset (or T1) (Onset of SDS Binding to the Dendrimer) and Csat (or T2) (Saturation of the Dendrimer by SDS) Values for (G3-G6)-PAMAM-NH2, (G3-G6)-PAMAM-OH, and (G3-G5)-PAMAM-Sugar-Ball Dendrimers from EMF and ITC Experiments at 298 Ka PAMAM dendrimer (0.1 mM)

MM

G3-PAMAM-NH2 G3-PAMAM-OH G3-PAMAM-sugar ball G4-PAMAM-NH2 G4-PAMAM-OH G4-PAMAM-sugar ball G5-PAMAM-NH2 G5-PAMAM-OH G5-PAMAM-sugar ball G6-PAMAM-NH2 G6-PAMAM-OH

3256 3272 6104 6909 6941 12605 14215 14279 25607 28825 28951

% w/v

natural pH

Conset (or T1) (EMF) (10-3 mol dm-3)

Csat (or T2) (EMF) (10-3 mol dm-3)

Csat (or T2) (ITC) (10-3 mol dm-3)

(Csat(EMF) - m1)b (10-3 mol dm-3)

0.033 0.033 0.061 0.069 0.069 0.126 0.142 0.143 0.256 0.288 0.290

10.2 7.4 7.2 10.2 7.4 7.2 10.2 7.4 7.2 10.2 7.4

0.8 1.8 2 0.7 0.7 0.6 0.4 ∼0.3 0.5 not deduciblec not deduciblec

20 17 ∼22 23 20 ∼22 28 22 ∼25 30-40 >25

19 18 ∼18 22 22 ∼18 27 25 ∼20 >30 >26

13 10 15 18 13 15 22 16 19 25-35 19

a The dendrimer concentration used was 0.1 × 10-3 mol dm-3. b m is the SDS monomer concentration at the total SDS concentration 1 (Csat). c Very low concentration binding.

Figure 1. Graph of the electromotive force of the DS- selective electrode (relative to a bromide ion electrode) as a function of total SDS concentration for the (0) pure SDS aqueous solution and (2) SDS/0.1 × 10-3 mol dm-3 G3-PAMAM-NH2 aqueous solution. Solutions are doped with 10-4 mol dm-3 NaBr. T ) 298 K. Conset and Csat of binding; pH 10.2.

Figure 2. Graph of the electromotive force of the DS- selective electrode (relative to a bromide ion electrode) as a function of total SDS concentration for the (0) pure SDS aqueous solution and (2) SDS/0.1 × 10-3 mol dm-3 G3-PAMAM-OH aqueous solution. Solutions are doped with 10-4 mol dm-3 NaBr. T ) 298 K. Conset and Csat of binding; pH 7.4.

to determine this critical concentration.15b,c,31-33 The reason for this is the difference in selectivity and sensitivity between the two methods. For example, it is sometimes possible to detect a weak enthalpy change via ITC or a change in surface tension, which is outside the scope of the sensitivity of the SDS electrode, which has the advantage of high selectivity. Such effects were shown in our recent work on linear and star polymers.32 The EMF and ITC data at natural pH for all the dendrimers (Gx-PAMAM-NH2, Gx-PAMAM-OH, and Gx-PAMAM-sugar balls) studied in this work are shown in Figures 1-7. Only a selection of the EMF data is displayed (Figures 1-4); the rest is found in the Supporting Information (Supporting Information Figures A-E). The convergency of the EMF and ∆Hi values with and without the dendrimers signifying saturation of binding at Csat (or T2) in the EMF and ITC experiments agrees well for both methods, as shown in Figures 1-7 and summarized in Table 2. We consider that Csat (or T2)

corresponds to the formation of micelles most likely on the surface of the PAMAM dendrimers, because the interior is too hydrophobic for highly charged aggregates (low dielectric constant). This is confirmed below by SANS measurements. Our Csat results correspond to the saturation of dendrimers by SDS micelles, which is in agreement with the former studies by Tomalia et al.9 and Esumi et al.11,13 Csat increases with increasing generation, most prominently for PAMAM-NH2. On the other hand, a welldefined Conset (or T1) value of cooperative binding which is manifested in a sharp break is only seen in the EMF data of all the G3-G5 dendrimers (see, e.g., Figures 1-3 for G3 and Supporting Information Figures A-E for G4 and G5). From EMF measurements, we observe a decrease in Conset (T1) with increasing dendrimer generation for PAMAM-NH2, PAMAM-OH, and the sugar balls. The systematic dependence on dendrimer generation of Conset and Csat was observed for the first time in our present studies (see Table 2). With the exception of the low generation Gx-PAMAMOH and sugar-ball dendrimers, no obvious Conset (or T1) value was detected in the ITC experiments (see Figures 5-7). We observe a common trend for (G3-G5)PAMAM-X dendrimers but an additional binding at very low SDS concentrations for G6-PAMAM-NH2 or -OH, which hides Conset (T1) under a continuous binding curve

(31) Li, Y.; Ghoreishi, S. M.; Warr, J.; Bloor, D. M.; Holzwarth, J. F.; Wyn-Jones, E. Langmuir 2000, 16, 3093. (32) Couderc-Azouani, S.; Sidhu, J.; Georgiou, T. K.; Charalambous, D. C.; Vamvakaki, M.; Patrickios, C. S.; Bloor, D. M.; Penfold, J.; Holzwarth, J. F.; Wyn-Jones, E. Langmuir 2004, 20, 6458. (33) Li, Y.; Ghoreishi, S. M.; Warr, J.; Bloor, D. M.; Holzwarth, J. F.; Wyn-Jones, E. Langmuir 1999, 15, 6326.

9324

Langmuir, Vol. 20, No. 21, 2004

Sidhu et al.

Figure 3. Graph of the electromotive force of the DS- selective electrode (relative to a bromide ion electrode) as a function of total SDS concentration for the (0) pure SDS aqueous solution and (2) SDS/0.1 × 10-3 mol dm-3 G3-PAMAM-sugar-ball aqueous solution. Solutions are doped with 10-4 mol dm-3 NaBr. T ) 298 K. Conset and Csat of binding; pH 7.2.

(Figure 4); therefore Conset (T1) could not be determined. Csat (T2) is still clearly identifiable in both the ITC and EMF data. This continuous low concentration binding is likely due to the larger number of surface groups, because the compact interior structure found by Tomalia et al.8 should not favor host-guest interactions between charged surfactants and the interior of G6-PAMAM-X dendrimers at pHs well above the pKs. The negative enthalpy values measured by ITC suggest electrostatic attraction. The EMF Conset (or T1) values are listed in Table 2 together with the binding capacity of the dendrimers as expressed at the binding limit by (Csat (or T2) - m1), with m1 being the monomer concentration of SDS at the various Csat (or T2) values; the latter was also measured by ITC. Since the dendrimers have well-defined molecular weights, we confined the present measurements to the same molar concentration of dendrimers, that is, 10-4 mol dm-3. Most studies of nonionic polymer/surfactant systems are conducted in the % w/v value of the polymer because of the uncertainty in the molecular mass. We therefore also listed the % w/v values in Table 2. One of the consequences of using the molar concentration is that the % w/v value of the dendrimers approximately doubles as the generation increases by one. The following conclusions can be drawn from the data on the (G3-G5)-PAMAM dendrimers. (1) The Conset (or T1) value or critical aggregation concentration (cac) which is clearly observed in the EMF data for the G3-G5 dendrimers is essentially a measure of the critical micelle concentration of the bound SDS micelles. These values are always much lower than the cmc of pure SDS, and the difference is attributed to the increased stability of the bound SDS micelles as a result of the reduction in their headgroup repulsion and other additional hydrophobic effects arising from the interaction of the polymer with the micellar surface. In all cases, Conset (or T1) decreases from G3 to G5, showing that the affinity of the dendrimers toward SDS micelles increases as the generation increases. Esumi et al.13 investigated the influence of the generation on the binding of anionic surfactants to sugar-persubstituted dendrimers at low surfactant concentrations using surface tension. The dendrimer/SDS complexes showed higher surface activities than SDS alone, leading to the assumption that the

Figure 4. (a) Graph of the electromotive force of the DSselective electrode (relative to a bromide ion electrode) as a function of total SDS concentration for the (0) pure SDS aqueous solution and (2) SDS/0.1 × 10-3 mol dm-3 G6-PAMAM-NH2 aqueous solution. Solutions are doped with 10-4 mol dm-3 NaBr. T ) 298 K. Csat of binding; pH 10.2 (natural pH). (b) Graph of the electromotive force of the DS- selective electrode (relative to a bromide ion electrode) as a function of total SDS concentration for the (0) pure SDS aqueous solution and (2) SDS/0.1 × 10-3 mol dm-3 G6-PAMAM-OH aqueous solution. Solutions are doped with 10-4 mol dm-3 NaBr. T ) 298 K. Csat of binding; pH 7.4 (natural pH).

tail is oriented toward air. In bulk solutions, they proposed that the dendrimer/SDS complexes formed a kind of network. This is hard to believe for us because Conset (or T1) is associated with a cooperative binding and SANS measurements (see below) showed SDS micellar aggregates on the surface of the dendrimers around Csat (or T2) but no further aggregation. (2) The Csat (or T2) values or binding capacities of the (G3-G5)-PAMAM-OH and sugar-ball dendrimers are very similar, suggesting that these functional surface groups have very little effect on the capacity of the dendrimers to bind SDS. On the other hand, there is a progressive increase with generation in the binding capacity of the Gx-PAMAM-NH2 dendrimers, suggesting that the NH2 groups play a significant role in the binding process and also show some recognition for the micellar surface. Surprisingly, the Csat (T2) values measured by the EMF and ITC experiments for other dendrimers were also in the range of 2 × 10-2 mol dm-3 SDS.14 This

Interactions of PAMAM Dendrimers with Surfactants

Figure 5. ITC graph of the enthalpy per injection as a function of total SDS concentration for (O) pure SDS aqueous solution, (b) SDS/0.1 × 10-3 mol dm-3 G3-PAMAM-NH2 aqueous solution, (2) SDS/0.1 × 10-3 mol dm-3 G4-PAMAM-NH2 aqueous solution, (9) SDS/0.1 × 10-3 mol dm-3 G5-PAMAMNH2 aqueous solution, and (/) SDS/0.1 × 10-3 mol dm-3 G6PAMAM-NH2 aqueous solution. T ) 298 K. Csat of binding; pH 10.2.

Figure 6. ITC graph of the enthalpy per injection as a function of total SDS concentration for (O) pure SDS aqueous solution, (b) SDS/0.1 × 10-3 mol dm-3 G3-PAMAM-OH aqueous solution, (2) SDS/0.1 × 10-3 mol dm-3 G4-PAMAM-OH aqueous solution, (9) SDS/0.1 × 10-3 mol dm-3 G5-PAMAM-OH aqueous solution, and (/) SDS/0.1 × 10-3 mol dm-3 G6-PAMAMOH aqueous solution. T ) 298 K. Csat of binding; pH 7.4.

emphasizes the role of the dendrimer structure in the binding process. (3) The observation of a clear Conset (or T1) value in the EMF studies indicates the formation of bound SDS micelles on the dendrimer surface, which was also found by Tomalia9,10 and Esumi11,13. The absence of a clear indication of Conset (or T1) in the ITC data suggests that other processes sensitive to enthalpy changes are hiding Conset (or T1). The difference between the ∆Hi values with and without the dendrimer at very low concentrations suggests that some binding occurs at concentrations lower than the critical aggregation concentration or Conset (or T1). The above data also indicate that the surface groups are dominating the binding process. The binding in the G6-PAMAM-NH2/SDS and G6PAMAM-OH/SDS systems is significantly different compared to that of the earlier generations in the sense that no onset of cooperative binding (Conset (or T1)) is observed either in the EMF or in the ITC data. In addition, the ITC experiments show that the difference between the corresponding ∆Hi values with and without the macromol-

Langmuir, Vol. 20, No. 21, 2004 9325

Figure 7. ITC graph of the enthalpy per injection as a function of total SDS concentration for (O) pure SDS aqueous solution, (b) SDS/0.1 × 10-3 mol dm-3 G3-PAMAM-sugar-ball aqueous solution, (2) SDS/0.1 × 10-3 mol dm-3 G4-PAMAM-sugarball aqueous solution, and (9) SDS/0.1 × 10-3 mol dm-3 G5PAMAM-sugar-ball aqueous solution. T ) 298 K. Csat of binding; pH 7.2.

Figure 8. ITC graph of the enthalpy per injection as a function of total SDS concentration in the presence of 0.1 mol dm-3 NaCl for (O) pure SDS aqueous solution, (b) SDS/0.1 × 10-3 mol dm-3 G3-PAMAM-NH2 aqueous solution, (2) SDS/0.1 × 10-3 mol dm-3 G4-PAMAM-NH2 aqueous solution, (9) SDS/0.1 × 10-3 mol dm-3 G5-PAMAM-NH2 aqueous solution, and (/) SDS/0.1 × 10-3 mol dm-3 G6-PAMAM-NH2 aqueous solution. T ) 298 K. Csat of binding; pH 10.2.

ecules increases as the SDS concentration is decreased. In the EMF data on G6-PAMAM-NH2 (Figure 4a), there might be two modes of binding with a transition around ∼0.4 × 10-3 mol dm-3, which could be the critical aggregation concentration (cac) or Conset (or T1) for the formation of bound SDS micelles, but this is uncertain. In the data on G6-PAMAM-OH (Figure 4b), there is no clear break point which could identify Conset. Clearly, in both of these dendrimers, the binding in excess of 10-3 mol dm-3 is accompanied by the formation of bound SDS micelles. At concentrations below 0.4 × 10-3 mol dm-3, the amount of bound SDS molecules per dendrimer varies from one to five, which indicates that the bound SDS is unlikely to be aggregated. In addition, the ∆Hi values are negative at low SDS concentrations, which suggests that here the binding mechanism could involve electrostatic attraction between the dendrimers and SDS. In an attempt to make further progress in this direction, we tried to shield electrostatic sites on the dendrimers by carrying out measurements in the presence of NaCl. The ITC data for the dendrimers in the presence of 0.1 M NaCl are shown in Figures 8-10.

9326

Langmuir, Vol. 20, No. 21, 2004

Sidhu et al.

Figure 9. ITC graph of the enthalpy per injection as a function of total SDS concentration in the presence of 0.1 mol dm-3 NaCl for (O) pure SDS aqueous solution, (b) SDS/0.1 × 10-3 mol dm-3 G3-PAMAM-OH aqueous solution, (2) SDS/0.1 × 10-3 mol dm-3 G4-PAMAM-OH aqueous solution, (9) SDS/0.1 × 10-3 mol dm-3 G5-PAMAM-OH aqueous solution, and (/) SDS/ 0.1 × 10-3 mol dm-3 G6-PAMAM-OH aqueous solution. T ) 298 K. Csat of binding; pH 7.4.

Figure 11. SANS scattered intensity for (O) 0.1 mM G4PAMAM-NH2/20 mM SDS, (∇) 0.72 mM (0.5% w/v) G4PAMAM-NH2/33 mM SDS, and (∆) 0.1 mM G4-PAMAMOH/19 mM SDS in D2O. The solid lines are model fits as described in the text, made by using eqs 1 and 2 and the parameters summarized in Table 3. T ) 298 K.

Figure 10. Graph of the enthalpy per injection as a function of total SDS concentration in the presence of 0.1 mol dm-3 NaCl for (O) pure SDS aqueous solution, (b) SDS/0.1 × 10-3 mol dm-3 G3-PAMAM-sugar-ball aqueous solution, (2) SDS/0.1 × 10-3 mol dm-3 G4-PAMAM-sugar-ball aqueous solution, and (9) SDS/0.1 × 10-3 mol dm-3 G5-PAMAM-sugar-ball aqueous solution. T ) 298 K. Csat of binding; pH 7.2.

Figure 12. SANS scattered intensity for (O) 0.17 mM (0.5% w/v) G6-PAMAM-NH2/77 mM SDS, (3) 0.5% w/v G6PAMAM-NH2/19 mM SDS, and (4) 0.1 mM G6-PAMAMNH2/26 mM SDS in D2O. The solid lines are model fits as described in the text, made by using eqs 1 and 2 and the parameters summarized in Table 3. T ) 298 K.

No EMF data were taken because of problems with the electrode in the presence of salt. All the ITC data in these diagrams show very similar behavior with a definite systematic trend in the ∆Hi data to more negative values as the dendrimer generation increases. Similar results showing negative ∆Hi values upon the addition of salt were also observed in the same SDS concentration range for poly(1,4-diaminobutane) (or DAB) dendrimer/SDS mixtures.14 Another important feature is that for each family of dendrimers the ∆Hi value for each titration with and without the dendrimer now decreases as the SDS concentration decreases, suggesting that by extrapolation to very low SDS concentrations the curves will eventually merge. (This could not be measured because of the ITC sensitivity limit.) Such a behavior was also observed in our former study on DAB dendrimer/SDS mixtures.14 The Conset (or T1) value observed in the presence of salt will decrease because the headgroup repulsions in the bound micelles will be further reduced by the extra bound Na+ counterions; this effectively pushes the Conset (or T1) value

to lower SDS concentrations. The merge of the ITC curves at very low SDS concentrations could not be fully observed because of very weak signals. In the case of the G6-PAMAM-OH and G6-PAMAMNH2 dendrimers, the addition of salt changes their ITC binding behavior and they all fall into a systematic trend for each family of dendrimers. ITC experiments were carried out to investigate whether dodecyltrimethylammonium bromide (DTAB) and also hexaethylene glycoln-dodecyl ether (C12EO6) interact with these dendrimers. In all cases studied, we did not observe any significant evidence inside experimental error for binding (see Supporting Information Figures F and G). The same result was observed in our former studies on DAB dendrimers with TTAB or C12EO6.14 Discussion of SANS Results. The SANS data for generations 3, 4, 5, and 6 of poly(amidoamine) dendrimers in combination with SDS are shown in Figures 11 and 12 and Supporting Information Figures H and I. The scattering from dendrimers only is negligible and is not

Interactions of PAMAM Dendrimers with Surfactants

Langmuir, Vol. 20, No. 21, 2004 9327

Table 3. Aggregation Numbers and Other Parameters (Surface Charge (Zed), the Radii of the Inner Core and the Outer Shell (R1 and R2, Respectively), the Effective Concentration (Ceff), and the Scale Factor (sf)) for SDS/Dendrimer Systems Derived from eqs 1 and 2 Are Related to SANS Measurementsa system

Nagg

Zed

R1b (Å)

R2 (Å)

Ceff (mM)

sf

0.1 mM G3-PAMAM-NH2 + 17 mM SDS (Csat) 0.1 mM G4-PAMAM-NH2 + 20 mM SDS (Csat) 0.72 mM (0.5% w/v) G4-PAMAM-NH2 + 33 mM SDS (Csat) 0.35 mM (0.5% w/v) G5-PAMAM-NH2 + 28 mM SDS 0.17 mM (0.5% w/v) G6-PAMAM-NH2 + 19 mM SDS (Csat) 0.17 mM (0.5% w/v) G6-PAMAM-NH2 + 26 mM SDS (>Csat) 0.17 mM (0.5% w/v) G6-PAMAM-NH2 + 77 mM SDS (.Csat) 0.1 mM G3-PAMAM-OH + 17 mM SDS (Csat) 0.1 mM G4-PAMAM-OH + 19 mM SDS (Csat) 0.1 mM G5-PAMAM-OH + 22 mM SDS (Csat) 0.1 mM G6-PAMAM-OH + 22 mM SDS (Csat)

64 67 65 72 79 75 78 90 108 73 58 61 66 92

9 10 15 19 11 16 13 19 17 26 10 11 13 14

16.7 16.7 16.7 16.7 16.7 16.7 16.7 16.7 16.7 16.7 16.7 16.7 16.7 16.7

20.2 20.5 20.3 21.4 21.7 21.3 21.6 22.2 24.1 21.1 19.6 19.9 20.4 22.8

12 18 28 80 20 49 28 40 35 100 11 15 18 18

1.04 1.03 1.03 0.90 1.01 0.92 1.05 0.8 0.93 0.7 1.03 1.1 1.06 1.01

a

T ) 298 K. b R1 is fixed by the model as a fully extended C12 chain length of the dodecyl sulfate molecule. Table 4. Effective Concentrations, Aggregation Numbers of SDS Micelles, and Structures of the SDS Micelle (M)/ Dendrimer (D) Complex for Different SDS/Dendrimer Concentrations Deduced from SANS Analysis at 298 Ka system

(C - m1) (mM)

Ceff (mM)

Nagg (SDS)

0.1 mM G3-PAMAM-NH2 + 17 mM SDS (Csat) 0.1 mM G4-PAMAM-NH2 + 20 mM SDS (Csat) 0.72 mM (0.5% w/v) G4-PAMAM-NH2 + 33 mM SDS (