3854
Langmuir 2004, 20, 3854-3862
Articles Partitioning of Macrocyclic Compounds in a Cationic and an Anionic Micellar Solution: A Small-Angle Neutron Scattering Study Eugenio Caponetti,* Delia Chillura-Martino, and Lucia Pedone Dipartimento di Chimica-Fisica “F. Accascina”, Universita` di Palermo, Viale delle Scienze Parco D’orleans II, Pad. 17, I-90128 Palermo, Italy Received June 17, 2003. In Final Form: January 30, 2004 Following a previous investigation on partitioning of some macrocycle compounds in sodium dodecyl sulfate (SDS) and dodecyltrimethylammonium bromide (DTAB) aqueous solutions and their effect on the micellar structure, a small-angle neutron scattering (SANS) study has been performed at fixed surfactant content (0.20 mol/L) and varying macrocycle concentrations from 0.20 up to 1.0 mol/L. Conductivity measurements have been also performed in order to evaluate the effect of the presence of macrocycles on the critical micellar concentration (cmc) of the two surfactants. SANS experimental data were fitted successfully by means of a core-plus-shell monodisperse prolate ellipsoid model. It has been found that 1,4,7,10,13,16-esaoxacyclooctadecane (18C6) and 4,7,13,16-tetraoxa-1,10-diazacyclooctadecane (22) do not interact with DTAB micelles whereas their sodium complexes interact with SDS aggregates and partially localize, as a consequence of electrostatic interaction, on the micellar surface or in the Stern layer. 2,5,8,11,14,17-Hexaoxabicyclo[16.4.0]dicosane (B18C6), as a consequence of the increased hydrophobic character with respect to 18C6, interacts with DTAB hydrocarbon chains and partially localizes in the inner part of micelles. This finding has been successfully used to justify the higher amount of B18C6 compared to the 18C6 one found in the SDS micellar phase. The substituted crown ether has been found localized both on the micelle surface via complex formation and in the inner part of micelles as a consequence of the increased hydrophobic character. For all systems, the aggregate size primarily decreases with the amount of macrocycle in the micellar phase. The interpretation of cmc trends as a function of macrocycle concentration gives information on its distribution between micellar and aqueous phases that is in line with SANS results.
Introduction Structural investigations on aqueous solutions of a single surfactant have been performed by using various techniques.1-4 In many technological applications, the addition of other components is requested to enhance some properties of the systems. For this reason, the effect of a third component has been an object of interest.5-9 It has been proved that the addition of several kinds of molecules strongly affects the structure of micellar aggregates.6,8 It is commonly accepted that highly hydrophobic species tend to locate themselves in the aggregate hydrophobic interiors. Polar or ionic species tend either to localize in the hydrophilic shell of the aggregates, directly affecting the * Corresponding author. E-mail:
[email protected]. Phone: ++39 091 6459842. Fax: ++39 091 590015. (1) Ikeda, S.; Hayashi, S.; Imae, T. J. Phys. Chem. 1981, 85, 106. (2) Missels, P. J.; Mazer, N. A.; Benedek, G. B.; Carey, M. C. J. Phys. Chem. 1983, 87, 1264. (3) Caponetti, E.; Triolo, R. Industrial and technological application of neutrons; Rustichelli, F.; Fontana, M.; Coppola, R., Eds.; NorthHolland: Amsterdam, The Netherlands, 1992; pp 403-424. (4) Asakawa, T.; Hisamatsu, H.; Miyagishi, S. Langmuir 1996, 12, 1193. (5) Muto, Y.; Yoda, K.; Yoshida, N.; Esumi, K.; Meguro, K.; BinanaLimbele, W.; Zana, R. J. Colloid Interface Sci. 1989, 130, 165. (6) Caponetti, E.; Chillura Martino, D.; Floriano, M. A.; Triolo, R. Langmuir 1997, 13, 3277. (7) Almgren, M.; Swarup, S. J. Phys. Chem. 1983, 87, 876. (8) Caponetti, E.; Causi, S.; De Lisi, R.; Floriano, M. A.; Milioto, S.; Triolo, R. J. Phys. Chem. 1992, 96, 4950. (9) Stilbs, P. J. Colloid Interface Sci. 1982, 87, 385; 1983, 94, 463.
structure, or to stay in the aqueous phase affecting the micelle structure by a variation of the solvent properties, by the so-called “solvent effect”.10 Macrocyclic compounds are particularly interesting because, as a consequence of strong ion-dipole interactions, they can form stable complexes with metal ions. The complex stability primarily depends on the matching between the ring dimension and the ion diameter. This property along with the selectivity shown by certain cyclic polyethers and polyamines toward cations is well documented11,12 and constitutes one of the interesting features which distinguish them from most noncyclic ligands. Macrocycles have been widely used in many areas where the above properties are important, such as in the fields of environmental chemistry, food industries, and pharmacology.11,12 When they are used along with surfactants, due to the contemporary presence of hydrophilic and hydrophobic portions, they can interact with micelles both by hydrophobic and hydrophilic interaction. In addition, in the case of anionic surfactants, it has been shown that macrocycles, being able to form complexes with surfactant counterions, are partially localized on the micellar surface and in the Stern layer.13-17 (10) Enea, O.; Jolicoeur, C. J. Phys. Chem. 1982, 86, 3870. (11) Izatt, R. M.; Pawlak, K.; Bradshaw, J. S.; Bruening, R. L. Chem. Rev. 1991, 1721. (12) Gokel, G. Crown Ethers & Criptands; The Royal Society of Chemistry, Cambridge, 1991.
10.1021/la035068h CCC: $27.50 © 2004 American Chemical Society Published on Web 04/13/2004
Partitioning of Macrocyclic Compounds
With the purpose to investigate the effects of some macrocyclic compounds on micelle structure and their partition between aqueous and micellar phases, a systematic investigation on dodecyltrimethylammonium bromide (DTAB) and sodium dodecyl sulfate (SDS) aqueous solutions containing alternatively different macrocyclic compounds has been undertaken by the small-angle neutron scattering (SANS) technique. This technique has been widely used in the study of hydrogen-containing twophase systems such as micellar aqueous solutions, where it represents a powerful technique because of the large difference between hydrogen and deuterium coherent scattering length densities. Besides aggregate size and shape determination, it allows the contemporary evaluation of aggregate interactions. In a first paper, the effect of the 1,4,7,10,13,16hexaoxacyclooctadecane crown ether on both SDS and DTAB solutions has been explored.18 Following Izatt nomenclature,11 the name of this crown ether is 18C6. The study was performed at fixed crown ether content and varying surfactant concentration. It has been shown19 that for such three-component micellar systems, an appropriate analysis of SANS experimental data can provide information either on modifications induced on micellar dimension, shape, and charge and on the additive partition between aqueous and micellar phases. In addition, some information on macrocycle localization inside the aggregates has been obtained. In particular, it was found that the 18C6 slightly affects the DTAB micelle structure and locates in the continuous phase slightly altering the solvent properties. This result excludes that interactions take place between crown ether molecules and the hydrophobic micelle core. On the other side, it has been found that the structure of SDS micelles is strongly affected by the presence of 18C6; this finding has been interpreted as due to the interaction between micelles and the sodium counterions complexed by macrocycle molecules. The proposed analysis of SANS data has been, subsequently, tested studying DTAB and SDS aqueous solutions containing alternatively equimolecular amounts of 18C6, 4,7,13,16-tetraoxa-1,10-diazacyclooctadecane (22), 4,7,13,16,21-pentaoxa-1,10-diazabicyclo[8.8.5]tricosane (221), 2,5,8,11,14,17-hexaoxabicyclo[16.4.0]dicosane (B18C6), and 2,5,8,15,18,21-hexaoxatricyclo[20.4.0.09.14]esacosane (Cy218C6).19 Names in parentheses follow the Izatt nomenclature.11 In the case of two DTAB systems, there was evidence for a prevalent localization of solute in the aqueous phase, whereas for the remaining DTAB and all the SDS systems it has been found that different, but in all cases appreciable, amounts of solute are localized in the micellar phase. The investigation on the macrocycle-surfactant systems was recently extended at fixed SDS content (0.20 mol/L) and varying macrocycle concentration in a range (13) Evans, D. F.; Sen, R.; Warr, G. G. J. Phys. Chem. 1986, 90, 5500. Evans, D. F.; Evans, J. B.; Sen, R.; Warr, G. G. J. Phys. Chem. 1988, 92, 784. (14) Baglioni, P.; Kevan, L. J. Chem. Soc., Faraday Trans. 1 1988, 84, 467. (15) Baglioni, P.; Rivara-Minten, E.; Kevan, L. J. Phys. Chem. 1988, 92, 4726. (16) Evans, D. F.; Sen, R.; Warr, G. G. J. Phys. Chem. 1986, 90, 5500. (17) Gould, I. R.; Kuo, P.-L.; Turro, N. J. J. Phys. Chem. 1985, 89, 3030. (18) Caponetti, E.; Chillura Martino, D.; Floriano, M. A.; Triolo, R.; Wignall, G. D. Langmuir 1995, 11, 2464. (19) Caponetti, E.; Chillura Martino, D.; Floriano, M. A.; Triolo, R. J. Mol. Struct. 1996, 383, 133.
Langmuir, Vol. 20, No. 10, 2004 3855
of 0-0.18 mol/L.20 It was found that the amount of B18C6 and Cy218C6 is larger than that of 18C6 and 22. This finding supports the hypothesis that 18C6 and 22 interact only with the charged surface of SDS micelles via complex formation between sodium ions and the macrocycles; B18C6 and Cy218C6, as a consequence of the presence of hydrophobic substituents, interact both with the micelle surface and with the hydrophobic region inside the micelles. To verify the aforementioned hypothesis, in the present paper, SANS data related to 0.20 mol/L DTAB and SDS aqueous solutions containing alternatively macrocycle amounts ranging from 0.20 up to 1.0 mol/L are presented and discussed. In addition, to verify the prevision about the distribution of a third component in micellar solutions, based on critical micelle concentration (cmc) variation as a function of additive concentration,21,22 conductivity measurements have been performed as a function of surfactant concentration in the presence of various amounts of the four macrocycles. Experimental Section Materials. 1,4,7,10,13,16-Hexaoxacyclooctadecane, Sigma, 4,7,13,16-tetraoxa-1,10-diazacyclooctadecane, Merck, 2,5,8,11,14,17-hexaoxabicyclo[16.4.0]dicosane, and 2,5,8,15,18,21-hexaoxatricyclo[20.4.0.09.14]esacosane, Fluka, were used as received. SDS, Fluka, and DTAB, Sigma, were crystallized from ethanol and an ethanol-ethyl acetate mixture, respectively, and dried in a vacuum oven at 60 °C for 2 days. D2O (99.8% D) was an Aldrich product. Methods. Small-Angle Neutron Scattering. SANS samples were prepared by weight using D2O as the solvent, maintaining a constant (0.2 mol/L) micellized surfactant concentration (C cmc). Due to the solubility of Cy218C6 in a 0.20 mol/L surfactant solution, the SANS investigation has been performed for DTAB aqueous solutions only in the presence of 18C6, 22, and B18C6, at concentrations whose values are reported in Table 1, and for SDS aqueous solutions in the presence of 18C6, 22, B18C6, and Cy218C6, at concentrations whose values are reported in Table 2. Data were collected on the W.C. Koheler 30 m SANS facility23 at Oak Ridge National Laboratory (ORNL), Oak Ridge, TN, and on the time-of-flight SANS instrument LOQ24 at the Rutherford Appleton Laboratory (RAL), ISIS pulsed neutron source, Didcot, Oxford, U.K. For both sets of experiments, samples were contained in fused silica cells of 1 or 2 mm path length, depending on the hydrogenated compound content. The temperature of the cell holder was kept constant at 25.00 ( 0.05 °C by circulating fluid from an external bath. The ORNL data were collected by using a 64 × 64 cm2 area detector and cell element size of ≈1 cm2 placed at 1.79 m from the sample holder. This geometry gives a Q-range (the scattering vector Q ) 4π sin θ/λ with 2θ being the scattering angle) of 0.04 < Q < 0.32 Å-1. The neutron wavelength was 4.75 Å (∆λ/λ ≈ 5%). Data were corrected for detector efficiency, instrumental background, and solvent intensity. The net intensities were converted to absolute ((5%) differential cross sections per unit sample volume (in units of cm-1) by comparison with precalibrated secondary standards.25 RAL data were collected by using the main two-dimensional detector placed at a distance of 4.3 m from the sample holder. Experimental intensities were averaged around annular rings before correction for transmission, detector efficiency, and monitor response. After corrections for the wavelength dependence of (20) Chillura Martino, D.; Caponetti, E.; Pedone, L. J. Appl. Crystallogr. 2003, 36, 562. (21) Bakshi, M. S.; Crisantino, R.; De Lisi, R.; Milioto, S. Langmuir 1994, 10, 423. (22) De Lisi, R.; Turco Liveri, V.; Castagnolo, M.; Inglese, A. J. Solution Chem. 1986, 15, 23. (23) Koehler, W. C. Physica (Utrecht) 1986, 137B, 320. (24) Heenann, R. K.; Penfold, J.; King, S. M. J. Appl. Crystallogr. 1997, 30, 1140. (25) Wignall, G. D.; Bates, F. S. J. Appl. Crystallogr. 1986, 20, 28.
3856
Langmuir, Vol. 20, No. 10, 2004
Caponetti et al.
Table 1. Floating and Derived Parameters from the Fitting Procedure Described in the Text for the DTAB-D2O System (First Two Rows) and for the DTAB-D2O-Macrocycle Systemsa R 0.20 0.20
VO (nm3)
νS (monomers)
Z (u.e.)
β
10-10F1 (cm-2)
45.5 45.7
72.7(8) 73.6(6)
18.1(9) 18.6(7)
[DTAB] mol/L 0.25 1.31 -0.39 0.25 1.56(5) -0.39
10-10F2 (cm-2)
10-10Fs (cm-2)
thick (Å)
Anhyd
102η
χ
3.33 3.32
6.33 6.33
3.9 3.8
4.8 4.7
6.3 6.3
11 8
3.35 4.80 4.72
6.07 5.62 5.23
3.7 3.2 2.9
4.7 4.8 4.9
5.9 5.5 5.0
8 6 4
0.20 0.55 1.01
0.02(2) -0.03(2) -0.03(3)
40.9 35.5 31.9
65.7(7) 60.5(9) 56.1(8)
16.7(3) 14.6(2) 12.9(4)
[18C6] mol/L 0.25 1.19 -0.39 0.24 1.09 -0.39 0.23 1.01 -0.39
0.20 0.55 1.01
0.01(1) -0.06(6) -0.01(1)
42.1 38.6 33.7
67(1) 66(1) 59(1)
15.1(5) 13.3(2) 10.6(5)
0.23 0.20 0.18
-0.39 -0.39 -0.39
3.32 4.32 4.94
6.09 5.68 5.26
3.9 3.3 3.0
4.9 4.9 5.1
5.9 5.4 5.0
5 4 3
0.20 0.61 0.99
0.41(2) 0.32(1) 0.28(1)
33.1 23.6 19.9
39.3(6) 21.7(4) 15.6(2)
12.2(4) 7.5(3) 5.2(3)
[B18C6] mol/L 0.31 1.81(7) 0.13 0.35 2.1(1) 0.38 0.33 2.4(2) 0.49
3.24 3.19 3.20
6.12 5.77 5.43
2.3 1.4 1.1
4.5 4.3 4.3
8.7 11.4 13.3
7 6 4
[22] mol/L 1.20 1.18 1.07
a The DTAB concentration for all the samples investigated is 0.20 mol/L. The error in parentheses refers to the last digit of the floating parameter values. R ) macrocycle fraction in the micellar phase, VO ) overall micelle volume, νS ) surfactant aggregation number, Z ) micelle charge, β ) dissociation degree, 1 ) core axial ratio, F1 ) core scattering length density, F2 ) shell scattering length density, Fs ) solvent scattering length density, thick ) shell thickness, Anhyd ) number of water molecules per surfactant molecule, η ) volume fraction, χ ) weighted sum of residual.
Table 2. Floating and Derived Parameters from the Fitting Procedure Described in the Text for the SDS-D2O System (First Two Rows) and for the SDS-D2O-Macrocycle Systemsa R 0.21 0.21
VO (nm3)
νS (monomers)
Z (u.e.)
β
61.5 60.9
90.4(6) 90.2(4)
16.6(5) 16.6(3)
0.18 0.18
10-10F1 (cm-2)
10-10F2 (cm-2)
10-10Fs (cm-2)
thick (Å)
Anhyd
102η
χ
[SDS] mol/L 1.67 -0.39 1.59(2) -0.39
6.03 6.03
6.33 6.33
4.8 4.8
8.8 8.9
8.9 8.9
4 2
4.42 4.83 5.32
6.15 5.68 5.29
5.0 4.7 4.4
6.9 7.4 8.0
9.9 8.6 8.4
1 1 1
0.20 0.61 1.02
0.29(1) 0.08(1) 0.04(1)
51.9 45.3 40.5
71.2(6) 63.5(9) 58.1(7)
20.6(7) 17.1(5) 13.9(4)
[18C6] mol/L 0.29 1.28 -0.39 0.27 1.14 -0.39 0.24 1.05 -0.39
0.20 0.62 1.01
0.25(6) 0.14(1) 0.08(1)
54.9 46.9 46.8
75.7(5) 61.0(9) 60.0(8)
19.4(6) 13.3(3) 12.2(2)
0.26 0.22 0.20
-0.39 -0.39 -0.39
4.71 4.03 4.02
6.14 5.72 5.37
5.0 5.2 5.2
7.3 6.7 6.8
9.6 10.3 10.3
2 2 2
0.21 0.60 1.01
0.67(1) 0.39(1) 0.30(1)
60.6 47.4 38.8
71.5(4) 45.0(8) 33.5(8)
21.5(2) 8.5(2) 3.9(2)
[B18C6] mol/L 0.30 1.68(5) -0.10 0.19 1.8(1) 0.32 0.12 1.7(1) 0.43
4.35 4.38 4.65
6.19 5.83 5.43
4.0 2.4 1.6
6.9 7.5 8.1
10.1 13.1 14.8
1 3 2
0.21
0.80(1)
58.9
55.2(6)
10.8(3)
[Cy218C6] mol/L 0.20 2.5(1) 0.07
3.95
6.24
3.0
7.5
13.3
5
[22] mol/L 1.36 1.09 1.08
a
The SDS concentration for all the samples investigated is 0.21 mol/L. The error in parentheses refers to the last digit of the floating parameters values. The symbols have the same meaning as in Table 1.
sample transmission, simultaneous neutron diffraction data from wavelengths in the range of 2-10 Å were combined to give the scattering intensities in the 0.018-0.23 Å-1 Q range. Scattering from solvent blanks was subtracted from that of each sample. Details on the data reduction can be found elsewhere.26 The net intensities were converted to absolute differential scattering cross sections per unit sample volume, dΣ(Q)/dΩ, in units of cm-1 by comparison with precalibrated secondary standards. Conductimetry. Conductivity samples were prepared by weight using degassed conductivity degree water. For each cmc determination, the macrocycle concentration was kept constant at the values reported in Table 3. Due to the low solubility of Cy218C6 in water, the cmc determination could not be extended at macrocycle concentrations higher than 0.013 mol/L. The electrical resistance measurements were performed by a calibrated ac bridge27 at a frequency of 2 kHz by using the type of cell with unplatinized electrodes described by Dagget et al.28 at a temperature of 25.00 ( 0.01 °C controlled by an external circulating water bath. (26) King, S. M. Modern Techniques for Polymer Characterisation; Pethrick, R. A., Dawkins, J. V., Eds.; John Wiley & Sons Ltd: Chichester, U.K., 1999. (27) Janz, G. J.; McInyre, J. D. E. J. Electrochem. Soc. 1961, 108, 72.
Results SANS Measurements. Experimental data for all systems investigated are reported in Figures 1-4. Data related to the 0.21 mol/L Cy218C6 have been already published, but since the model here proposed has been implemented with respect to the one used in the previous paper, the data have been reanalyzed. The effect of macrocycle addition on the SANS patterns of the two surfactant systems differs with the chemical nature of the surfactant and the macrocycle. For all DTAB systems, with increasing macrocycle concentration, the intensity dramatically decreases. The interaction peak position is not affected by the presence of 18C6 and 22 (Figures 1A and 2A), whereas it moves toward higher Q values as a consequence of B18C6 addition (Figure 3A). In the SDS system, the addition of 18C6 and 22 causes an intensity decrease (Figures 1B and 2B) whereas the (28) Dagget, H. N.; Bair, E. S.; Kraus, C. A. J. Am. Chem. Soc. 1951, 73, 799.
Partitioning of Macrocyclic Compounds
Langmuir, Vol. 20, No. 10, 2004 3857
Table 3. Critical Micelle Concentration (cmc, mol/L) of SDS and DTAB and Dissociation Degree at the cmc (β, e.u/ Molecule) in the Presence of 18C6, 22 and B18C6 SDS [18C6]
103cmc
β
[22]
103cmc
β
[B18C6]
103cmc
β
[Cy218C6]
103cmc
β
0.007 0.015 0.022 0.062 0.099 0.137 0.201
6.82 6.09 5.60 5.20 4.99 4.90 4.85
0.44 0.43 0.44 0.46 0.42 0.50 0.45
0.007 0.015 0.025 0.050 0.100 0.145 0.198
6.89 5.94 5.40 5.23 5.18 5.18 5.13
0.58 0.76 0.78 0.72 0.79 0.80 0.86
0.007 0.015 0.025 0.050 0.100 0.149 0.199
4.22 3.72 3.29 2.99 2.75 2.55 2.42
0.46 0.45 0.46 0.47 0.56 0.64 0.67
0.0015 0.0025 0.005 0.010 0.013
4.69 4.35 4.14 3.82 3.71
0.78 0.77 0.85 0.86 0.94
[18C6]
103cmc
β
[22]
103cmc
β
[B18C6]
103cmc
β
[B18C6]
103cmc
β
0.025 0.050 0.100 0.149 0.203
15.5 15.8 16.0 16.3 16.4
0.25 0.27 0.28 0.27 0.29
0.025 0.050 0.099 0.150 0.199
15.4 16.3 16.3 16.3 16.4
0.26 0.27 0.28 0.28 0.27
0.01 0.025 0.051 0.100 0.150 0.200
14.0 13.1 12.1 10.9 10.1 10.2
0.36 0.31 0.34 0.42 0.51 0.59
0.0025 0.0050 0.010 0.013
14.3 14.1 13.7 13.6
0.27 0.28 0.30 0.32
DTAB
a The cmc’s of SDS and DTAB in pure water were 8.11 × 10-3 and 15.3 × 10-3 mol/L, and the corresponding β values were 0.41 and 0.24 e.u./molecule.
presence of B18C6 and Cy218C6 causes an intensity increase (Figures 3B and 4) at 0.20 mol/L, in line with previous observations at lower macrocycle concentration.20 For the B18C6, at higher concentration an intensity decrease is observed. In all systems, the peak position moves toward higher Q values, but the effect is much stronger for the B18C6-containing system (Figure 3B). The intensity decrease, on increasing macrocycle concentration, is consistent with a particle size decrease. The peak position displacement toward higher Q values, observed for B18C6-DTAB and all SDS systems, is consistent with both a decrease of aggregate size and an increase of the dispersed phase volume fraction, η, as a consequence of macrocycle partial localization in the micellar phase. Conductivity Measurements. A large number of conductivity measurements have been performed at various macrocyle contents, varying the two surfactant concentrations. In particular, data have been collected from 0 up to 0.025 and 0.015 mol/L for the DTAB and SDS, respectively. For all systems investigated, the observed trends are similar; therefore, only values for seven aqueous solutions containing fixed amounts of B18C6 as a function of DTAB and SDS concentration are shown in Figure 5A,B. The remaining data will be accessible on request. Data Analysis and Discussion SANS Data. To reproduce SANS experimental intensities, it is necessary to describe the system under investigation by using an appropriate physical model. On this basis, for a two-phase system like a micellar system, the scattering cross section can be calculated by the following expression:29,30
dΣ (Q)/dΩ ) NpP(Q)S(Q) - Np∆(Q) + C
(1)
where Np is the particle number density (in units of NAl-1, with NA being Avogadro’s number), P(Q) is the scattering function of the single particle equal to the square mean (29) Caponetti, E.; Triolo, R. Adv. Colloid Interface Sci. 1990, 32, 235. (30) Caponetti, E.; Floriano, M. A.; Varisco, M.; Triolo, R. Structure and dynamics of supramolecular aggregates and strongly interacting colloids; Chen, S.-H., Huang, J. S., Tartaglia, P., Eds.; Kluwer Academic: Dordrecht, 1992; p 403.
value of the form factor F(Q), S(Q) is the structure function, ∆(Q) ) 〈F(Q)2〉 - 〈F(Q)〉2 is a term that accounts for deviation from sphericity and/or from monodispersity, and C is a term that accounts for the incoherent scattering and instrumental background. For rigorously monodisperse spheres, ∆(Q) ) 0. P(Q) depends on the particle size and shape and on the resulting distribution of the atomic scattering length densities within the aggregate. S(Q) is related to the interparticle interactions and depends on the micellar phase volume fraction, η, on the dimension, and on the single particle net charge. Once a model for P(Q) and S(Q) is defined, the total intensity can be computed by means of eq 1. The agreement between computed and experimental intensities can be improved by allowing some quantities of physical interest to vary. Binary Systems. DTAB-D2O and SDS-D2O SANS data were modeled using the ellipsoidal “core-plus-shell” model used in ref 18; in this model the core contains the entire alkyl chain and the shell contains the charged headgroups, a fraction of counterions, and hydration water molecules; other models,31 like the three-shell model reported for SDS micelles or the dry-shell reported for DTAB, could be used, but the general conclusions would hardly change. 〈F(Q)2〉 and the 〈F(Q)〉2 terms were computed by averaging the scattering amplitude over all orientations of the aggregate with respect to the direction of the scattering vector. The scattering lengths for the various groups were computed from atomic scattering lengths32 and from volumes reported in the literature;33 both quantities are summarized in Table 4 together with the hydration numbers34 used in building up the micelle structural model. Full details on the calculation procedure can be found elsewhere.18 S(Q) was computed by means of the rescaled mean spherical approximation (RMSA) using a screened Coulombic potential plus hard sphere repulsion.35 (31) Beer, S. S. J. Phys. Chem. 1987, 91, 4760. Beer, S. S.; Caponetti, E.; Johnson, J. J., Jr.; Jones, R. R. M.; Magid, L. J. J. Phys. Chem. 1986, 90, 5766. (32) Bacon, G. E. Neutron Diffraction, 3rd ed.; Clarendon: Oxford, 1975; p 38. (33) Immirzi, A.; Perini, B. Acta Crystallogr., Sect A 1972, 33, 216. Millero, F. J. Water and Aqueous Solution; Horne, R. A., Ed.; WileyInterscience: New York, 1982; p 519. (34) Høiland, H.; Ringseth, J. A.; Brun, T. S. J. Solution Chem. 1979, 8, 779.
3858
Langmuir, Vol. 20, No. 10, 2004
Figure 1. Experimental SANS differential scattering cross section: (A) DTAB aqueous solutions at [DTAB] ) 0.20 mol/L in the presence of 18C6 at [18C6] ) 0.00 (filled squares), 0.20 (open squares), 0.55 (filled triangles), and 1.01 (open triangles) mol/L. (B) SDS aqueous solutions at [SDS] ) 0.21 mol/L in the presence of 18C6 at [18C6] ) 0.00 (filled squares), 0.20 (open squares), 0.61 (filled triangles), and 1.02 (open triangles) mol/ L. The solid lines represent the calculated intensities by using the model described in the text.
The adjustable parameters in the fit procedure were the micelle total net charge, Z, and the aggregation number, ν. The total micelle volume was computed from ν and from the volume of the micelle constituents. The minor semi-axis of the micelle core, a1, was fixed to the length of the fully extended alkyl chain; therefore the core axial ratio 1 was computed once the volume of the aggregates was determined. The shell thickness, thick, was assumed to be uniform in all directions; because the overall minor semi-axis is defined by a2 ) a1 + thick and the overall major semi-axis by 2a2 ) a11 + thick, it follows that 2, the total axial ratio, differs from 1. Np was derived from ν and from the surfactant stoichiometric concentration corrected by the cmc. The C term of eq 1 was evaluated at high Q values in the Porod region of the scattering curve from the slope of Q4 I(Q) versus Q4.36 By using this fitting procedure, the calculated intensities for the two surfactant aqueous solutions were in good (35) Ashcroft, N. W.; Lekner, J. Phys Rev. 1966, 145, 83. Hayter, J. B.; Penfold, J. Mol. Phys. 1982, 42, 109. J. Chem. Soc., Faraday Trans. 1 1981, 77, 1851. (36) Porod, G. Kolloid Z. 1951, 124, 831.
Caponetti et al.
Figure 2. Experimental SANS differential scattering cross section: (A) DTAB aqueous solutions at [DTAB] ) 0.20 mol/L in the presence of 22 at [22] ) 0.00 (filled squares), 0.20 (open squares), 0.55 (filled triangles), and 1.01 (open triangles) mol/ L. (B) SDS aqueous solutions at [SDS] ) 0.21 mol/L in the presence of 22 at [22] ) 0.00 (filled squares), 0.20 (open squares), 0.62 (filled triangles), and 1.01 (open triangles) mol/L. The solid lines represent the calculated intensities by using the model described in the text.
agreement with the experimental data (see Figures 1-4); values of the fitting parameters, along with some derived quantities, are reported in Tables 1 and 2 and are in good agreement with the previous ones.18,20 In some of the ternary system data analysis, we will use the axial ratio as an additional variable parameter for reasons that will be clear in the following. Therefore, the fit of the DTAB and SDS data has been repeated by allowing to vary and the results have been added to Tables 1 and 2. They do not substantially differ from the ones obtained by using the two-parameter fit, indicating that the model is reliable. Ternary Systems. The experimental intensities, related to ternary systems, were reproduced by means of the model used for the binary systems opportunely modified. To allow the presence of a certain amount of macrocycle in the aggregates, as has been already done,18,20 a parameter R, defined as the fraction of macrocycle in the micellar phase, has been used as an additional adjustable parameter in the minimization procedure. η values were derived once R and νS were known. For each
Partitioning of Macrocyclic Compounds
Langmuir, Vol. 20, No. 10, 2004 3859
Figure 4. Experimental SANS differential scattering cross section related to SDS aqueous solutions at [SDS] ) 0.21 mol/L in the presence of Cy218C6 at [Cy218C6] ) 0.00 (filled squares) and 0.21 (open squares) mol/L. The solid lines represent the calculated intensities by using the model described in the text.
Figure 3. Experimental SANS differential scattering cross section: (A) DTAB aqueous solutions at [DTAB] ) 0.20 mol/L in the presence of B18C6 at [B18C6] ) 0.00 (filled squares), 0.20 (open squares), 0.61 (filled triangles), and 0.99 (open triangles) mol/L. (B) SDS aqueous solutions at [SDS] ) 0.21 mol/L in the presence of B18C6 at [B18C6] ) 0.00 (filled squares), 0.21 (open squares), 0.60 (filled triangles), and 1.01 (open triangles) mol/L. The solid lines represent the calculated intensities by using the model described in the text.
system, the determined macrocycle amount has been localized according to both surfactant and macrocycle properties; furthermore, the comparison between results obtained from the cationic and anionic surfactants in the presence of the same macrocycle was determining. For the 18C6- and 22-DTAB systems, R values in the limit of experimental errors are zero, confirming that no interaction between surfactant and macrocycle takes place. This implies that the hydrophilic interactions dominate over the hydrophobic ones preventing any localization in micelles. Therefore, for the 18C6- and 22SDS systems, the macrocycle amount present in the micellar phase has been localized on the aggregate surface as sodium complex. The value of 1.3 for the hydration number of sodium complex was used, being plausible that the cation, to enter the crown ether hole, must lose part of its 6 hydration molecules.34 The calculated intensities, as can be seen in Figures 1 and 2, very well agree with the experimental data. Corresponding values for the fitting parameters and some derived quantities are reported in Tables 1 and 2.
For the B18C6-DTAB system, R values significantly differ from zero, indicating that the increased hydrophobic character of the crown is sufficient to promote its localization in the micellar phase. Since its presence at the micelle surface has been excluded, it has to be localized in the micellar core. Obviously, the constraint that one of the micelle core dimensions is equal to the length of the fully extended alkyl chain no more holds. Therefore, the core axial ratio has been used as an additional variable parameter; a1 has been computed once V1 and 1 were known. The B18C6 and Cy218C6 R values in the SDS micellar phase are higher compared to those obtained for 18C6 and 22. Since the four compounds investigated have similar values of complexation constant, it can be expected that for all of them the amount in micelles should be roughly the same. On this basis, for each composition the amount of B18C6 and Cy218C6 at the micelle surface has been set equal to the mean R value obtained for 18C6 and 22, that is, 0.27, 0.11, and 0.06, respectively. The remaining part of the R fraction has been localized in the core. Such an assumption is in agreement with the presence of B18C6 in the DTAB micellar phase. The resulting calculated intensities, shown in Figures 3 and 4, agree very well with the experimental data. Corresponding values for the fitting parameters and some derived quantities are reported in Tables 1 and 2. The R values are reported as a function of macrocycle concentration in Figure 6A,B for DTAB and SDS systems containing 18C6, 22, B18C6, and Cy218C6, respectively. In the same figure, values previously obtained at low macrocycle concentration for SDS systems20 are also reported. In all cases, R decreases with macrocycle concentration. Two different trends are observed. For the 18C6- and 22-SDS systems, R values are close to each other, in agreement with the similar hydrophobic balance and complexation constant. For the B18C6, the R values are higher because the B18C6 molecules are localized both in the core and in the shell. For each composition, if the amount located in the shell is subtracted from R, the obtained amount is close to that found in the DTAB micellar core. This finding increases the confidence in the
3860
Langmuir, Vol. 20, No. 10, 2004
Caponetti et al.
Figure 5. Specific conductivity vs surfactant concentration: (A) The B18C6-DTAB system. From upper to lower curve, the [B18C6] is 0.00, 0.01, 0.025, 0.051, 0.10, 0.15, and 0.20 mol/L. (B) The B18C6-SDS system. From upper to lower curve, the [B18C6] is 0.00, 0.007, 0.015, 0.025, 0.050, 0.10, 0.15, and 0.20 mol/L.
Figure 6. Fraction of macrocycle micelles, R, vs crown ether concentration. (A) DTAB aqueous solutions in the presence of 18C6 (filled squares), 22 (open squares), and B18C6 (filled triangles). (B) SDS aqueous solutions in the presence of 18C6 (filled squares), 22 (open squares), B18C6 (filled triangles), and Cy218C6 (open triangles).
Table 4. Group Parameters Used in Modeling Experimental SANS Dataa
Values previously obtained at low macrocycle concentration for SDS systems are also reported.20 A decrease is observed for all the systems investigated, but the strength of the effect depends on the chemical nature of the macrocycles. In particular, for the 18C6- and 22-SDS systems ν reduces its value up to 67% of the initial value at [macrocycle] ) 1.0 mol/L. For the B18C6-containing systems, the effect is more dramatic (38%). From the observed trends along with the R trends, it can be inferred that the presence of macrocycles destabilizes the micellar aggregates, this effect being higher when the macrocycle penetrates inside the micellar core. For the 18C6- and 22-DTAB systems, ν slightly decreases. Since no macrocycle has been found in the micellar phase, this finding can be attributed to the solvent effect.10 The ν reduction at about 25% of the initial value observed for the DTAB-B18C6 system confirms the higher destabilizing ability of the macrocycle. Conductivity Results. The specific conductivity values for the B18C6-DTAB and B18C6-SDS systems (see Figure 5A,B), at each macrocycle concentration, are linearly correlated to the surfactant concentration in both the premicellar and postmicellar regions. The intersection point between the two straight lines identifies the cmc value.21 The slope in the premicellar region is greater than
MW SDS DTAB 18C6 22 B18C6 Cy218C6 MNa+ CH3 CH2 SO4Na+ BrN(CH3)3+ D2O
288.4 308.4 264.3 262.4 312.2 372.5
b (10-12 cm)
V (Å3)
2.48 3.20 5.14 3.32
412.0 491.8 374.3 380.8 463.7 613.6
-0.46 -0.083 2.60 0.36 0.68 -0.43 1.91
54.3 26.9 57.9 3.94 39.3 102.3 30.2
10-10F (cm-2) hyd no.
0.49 0.64 1.11 0.54 -0.84 -0.31 4.49 9.21 1.73 -0.42 6.34
1.3 4.0 6.0 5.0 1.0
a b ) scattering length, ref 32; V ) group volume, refs 33; F ) i scattering length density; hyd no. ) hydration number, ref 34.
model. For the Cy218C6, the higher value, in line with values obtained at lower concentrations,20 could be explained considering the increased hydrophobic character of the compound. The DTAB and SDS ν values are reported as a function of macrocycle concentration in Figure 7A,B, respectively.
Partitioning of Macrocyclic Compounds
Langmuir, Vol. 20, No. 10, 2004 3861
Figure 8. cmc/cmc0 vs crown ether concentration for 18C6DTAB (filled squares), 22-DTAB (open squares), B18C6-DTAB (filled up-triangles), Cy218C6-DTAB (open up-triangles), 18C6SDS (filled circles), 22-SDS (open circles), B18C6-SDS (filled down-triangles), and Cy218C6-SDS (open down-triangles) systems.
Figure 7. Aggregation number vs crown ether concentration. (A) DTAB aqueous solutions in the presence of 18C6 (filled squares), 22 (open squares), and B18C6 (filled triangles). (B) SDS aqueous solutions in the presence of 18C6 (filled squares), 22 (open squares), B18C6 (filled triangles), and Cy218C6 (open triangles).
that in the postmicellar one because of the variation in both the number of conducting species from 2N monovalent ions (with N being the number of surfactant molecules) to N/ν micelles having a charge Z, plus Z(N/ν) counterions, and the corresponding ionic mobility. For the B18C6-DTAB system, in the premicellar region, the increase of macrocycle content does not affect the slope, whereas in the postmicellar region it causes a slope increase. This finding is in agreement with results obtained from SANS data analysis (see Table 1) indicating that, in the micelle, the surfactant dissociation degree increases. That means that both the micelle charge and the free counterion number increase. The increase of macrocycle content in both the premicellar and the postmicellar regions of the B18C6-SDS system causes a conductivity decrease. In the postmicellar region, this behavior can be attributed to the decreased dissociation degree of the surfactant in the micelle (see Table 2) and to the increased number of complexed sodium ions whose specific conductivity is lower than that of free sodium ions. In the premicellar region, only the latter effect holds. The cmc values obtained for all the examined systems, reported in Table 3, are shown in Figure 8 as a function of macrocycle concentration; to plot all results on the same
scale, the cmc values have been normalized to the ones in pure water. The slight cmc increase with macrocycle concentration for the 18C6-DTAB system is in agreement with results on the same system reported in a previous paper where it was claimed that the 18C6 is almost completely localized in the aqueous phase.21 On increasing 18C6 concentration, the cmc of SDS decreases, approaching a constant value, in agreement with previous findings for the 18C6-SDS system.21 A similar trend was observed for the 18C6SDS system and was explained considering the association of the crown ether as a complex to the SDS micelle surface.21 In both cases, only a qualitative discussion was performed. The cmc values of the 22-DTAB and 22-SDS systems follow trends similar to those of the 18C6-DTAB and 18C6-SDS systems. For the B18C6- and Cy218C6-DTAB systems, the cmc decreases. Similar trends obtained in the study of the SDS-18C6 system and of butanol- and pentanol-DTAB aqueous solutions have been interpreted as due to the presence of a consistent additive amount in the micellar phase.21,22 Therefore, in the present case, since the complex formation has been excluded, the macrocycles have to be localized in the core. The higher cmc decrease observed for the SDS systems containing either B18C6 or Cy218C6 can be explained considering a partial localization of the two compounds both in the core, as for the B18C6- and Cy218C6-DTAB systems, and in the shell, as for the 18C6- and 22-SDS systems. These findings are perfectly in agreement with SANS conclusions. The only difference is that from cmc trends we can perform only qualitative considerations while from SANS results quantitative information has been inferred. Conclusions SANS performed on SDS and DTAB aqueous solutions in the presence of four macrocyclic compounds provided information on the modifications induced on micellar aggregate structure and on the distribution of each macrocycle between the aqueous and the micellar phase.
3862
Langmuir, Vol. 20, No. 10, 2004
From a comparison between results related to the two surfactant systems containing the same macrocycle, it was possible to determine the localization of each macrocycle in the micellar phase. The crown ether 18C6 and cryptand 22 are distributed among SDS micelles and the solvent while they remain almost completely in the solvent when added to DTAB aqueous solutions. It follows that the presence of 18C6 and 22 in SDS micelles has to be attributed exclusively to electrostatic interaction between sodium-macrocycle complexes and the charged micellar surface. The B18C6 is distributed among the micellar and aqueous phases in both DTAB and SDS solutions. From these results, it follows that the higher amount of B18C6 in SDS micelles with respect to that of 18C6 and 22 is consistent with a partial localization in the micellar shell as a consequence of electrostatic interaction and a partial localization in the micellar core as a consequence of hydrophobic interaction.
Caponetti et al.
In all systems, the presence of macrocycles causes a general reduction of micellar aggregation number that follows the macrocycle amount localized in the micellar phase. The distributions between aqueous and micellar phases obtained from the cmc trends are in line with those obtained by SANS. Acknowledgment. We thank the Rutherford Appleton Laboratory (RAL) Committee for giving us the opportunity to perform small-angle neutron scattering measurements. Thanks are due to R. K. Heenann and S. King for technical assistance and useful discussion. We gratefully acknowledge financial support of this work by the Ministero dell’Istruzione, dell’Universita` e della Ricerca (MIUR), and the Consiglio Nazionale delle Ricerche (CNR). LA035068H
